{ "schemaVersion": "1.0", "updated": "2026-09-05", "ratingPolicy": "Editorial fit for the stated task, not a benchmark or global quality ranking. Null means not rated.", "resources": [ { "id": "resource-b5f039230c", "title": "How a form is built here", "domain": "UI", "domains": [ "UI" ], "kind": "Implementation recipe", "summary": "Six rules read on a live operator form: asterisks, numbered cards, a searchable list, an aligned help icon.", "href": "forms.html", "sourcePath": "forms.html", "thumbnail": "assets/thumbs/pages/resource-b5f039230c.jpg", "keywords": [ "forms.html", "form", "forms", "asterisk", "required", "combobox", "searchable", "list", "help", "icon", "cards", "rail", "numbered", "cards", "operator", "form", "space", "registry", "twin", "form", "forms", "asterisk", "required", "combobox", "searchable", "list", "help", "icon", "cards", "rail", "numbered", "cards", "quiet", "rail", "space", "registry", "twin", "operator", "form" ] }, { "id": "resource-9002d8c17d", "title": "Put a speaker in the page", "domain": "Speakers", "domains": [ "Speakers", "Voice", "UI" ], "kind": "Integration recipe", "summary": "A person who presents the page in a real voice.", "href": "people/index.html", "sourcePath": "people/index.html", "thumbnail": "assets/thumbs/pages/resource-9002d8c17d.jpg", "keywords": [ "people/index.html", "speaker", "valentina", "bust", "face", "half", "bust", "full", "body", "presenter", "avatar", "sizes", "speaker", "speakers", "sizes", "speaker", "sizes", "valentina", "zalia", "grace", "person", "bust", "face", "half", "bust", "full", "body", "presenter", "talking", "avatar", "cost", "still", "speaker", "moving", "speaker" ] }, { "id": "resource-9c544e774f", "title": "Make a film", "domain": "Video", "domains": [ "Video", "Voice", "Images" ], "kind": "Workflow", "summary": "A story to the screen, batch by batch.", "href": "film.html", "sourcePath": "film.html", "thumbnail": "assets/thumbs/pages/resource-9c544e774f.jpg", "keywords": [ "film.html", "film", "movie", "video", "story", "children", "book", "storyboard", "seedance", "series", "film", "movie", "video", "story", "children", "book", "storyboard", "seedance", "batches", "series", "animation", "prompts" ] }, { "id": "resource-03b113b3e2", "title": "Globes in dots", "domain": "Images", "domains": [ "Images" ], "kind": "Worked example", "summary": "Seven dotted-globe designs, with editable drawings and the method to reframe them.", "href": "diagramas/globes/index.html", "sourcePath": "diagramas/globes/index.html", "thumbnail": "assets/thumbs/cards/card-globes.svg", "keywords": [ "globe", "globes", "world", "map", "dots", "dotted", "continents", "earth", "planet", "watermark", "background", "pale", "grey", "behind", "text", "space", "registry", "cyprus", "europe", "africa", "americas", "asia", "sphere", "svg", "vector", "currentcolor", "globe", "globes", "world", "map", "dots", "dotted", "continents", "earth", "planet", "watermark", "background", "pale", "grey", "behind", "text", "space", "registry", "cyprus", "europe", "africa", "americas", "asia", "sphere", "svg", "vector", "currentcolor" ] }, { "id": "resource-45ce3da32d", "title": "Ways of presenting images", "domain": "Images", "domains": [ "Images", "UI" ], "kind": "Technique collection", "summary": "Tunnels, globes, spirals, grids and stacks: photographed patterns for presenting an image collection.", "href": "inspiration/presenting-images.html", "sourcePath": "inspiration/presenting-images.html", "thumbnail": "assets/thumbs/animos/card-tunnel.jpg", "keywords": [ "presenting", "images", "ways", "of", "presenting", "images", "show", "presentation", "of", "pictures", "screens", "card", "tunnel", "spiral", "globe", "sphere", "ring", "wheel", "carousel", "orbit", "grid", "mosaic", "stack", "deck", "cascade", "totem", "ticker", "film", "strip", "animos", "3d", "slideshow", "diaporama", "moving", "pictures", "product", "shots", "portfolio", "reel", "show", "my", "screens", "in", "motion", "presenting", "images", "ways", "of", "presenting", "images", "show", "presentation", "of", "pictures", "screens", "card", "tunnel", "spiral", "globe", "sphere", "ring", "wheel", "carousel", "orbit", "grid", "mosaic", "stack", "deck", "cascade", "totem", "ticker", "film", "strip", "animos", "3d", "slideshow", "diaporama", "moving", "pictures", "product", "shots", "portfolio", "reel", "show", "my", "screens", "in", "motion" ] }, { "id": "resource-ec9a9105a6", "title": "Illustrate a concept", "domain": "Images", "domains": [ "Images" ], "kind": "Workflow recipe", "summary": "2D and 3D forms that say the idea before words: Earth, stars, clay, paper.", "href": "diagramas/index.html", "sourcePath": "diagramas/index.html", "thumbnail": "diagramas/examples/d-tierra-red.png", "keywords": [ "diagramas/index.html", "diagram", "diagrama", "concept", "illustration", "shapes", "orbit", "strata", "delta", "bridge", "diagram", "diagrams", "concept", "illustration", "shapes", "2d", "3d", "earth", "stars", "constellation", "orbit", "strata", "delta", "bridge", "materials", "structures", "no", "text" ] }, { "id": "resource-2249826e42", "title": "Putting it online", "domain": "Workflows", "domains": [ "Workflows", "Voice", "Images", "UI" ], "kind": "Publishing recipe", "summary": "Three roads to a link, and the one question that picks between them.", "href": "publishing/index.html", "sourcePath": "publishing/index.html", "thumbnail": "assets/thumbs/pages/resource-2249826e42.jpg", "keywords": [ "publishing/index.html", "publish", "share", "url", "slug", "deploy", "louis", "hetzner", "sandbox", "private", "link", "publish", "publishing", "put", "online", "share", "link", "url", "share.eregistrations.dev", "slug", "update", "in", "place", "studio", "publish.sh", "deploy", "louis", "hetzner", "server", "branded", "domain", "download", "sandbox", "private", "link", "send", "this", "to", "someone" ] }, { "id": "resource-b5a76afa99", "title": "Our page design rules", "domain": "UI", "domains": [ "UI" ], "kind": "Design standard", "summary": "Type, spacing, colour and every block, each one shown on the page.", "href": "documentation-compendium.html", "sourcePath": "documentation-compendium.html", "thumbnail": "assets/thumbs/pages/resource-b5a76afa99.jpg", "keywords": [ "documentation-compendium.html", "kennedy", "teletubbies", "compendium", "tokens", "typography", "hierarchy", "blocks", "callout", "design", "rules", "typography", "spacing", "color", "colour", "tokens", "kennedy", "teletubbies", "layout", "page", "html", "glance", "document", "triple", "read", "block", "vocabulary", "reading", "measure" ] }, { "id": "resource-01bccbce86", "title": "The ready document formats", "domain": "UI", "domains": [ "UI" ], "kind": "Template collection", "summary": "The page shapes to copy: leaflet, teaching page, plan, poster, deck.", "href": "formats/index.html", "sourcePath": "formats/index.html", "thumbnail": "assets/thumbs/pages/resource-01bccbce86.jpg", "keywords": [ "formats/index.html", "format", "template", "exemplar", "one-pager", "vision", "paper", "concept", "note", "proposal", "formats", "format", "leaflet", "teaching", "page", "explainer", "plan", "deck", "one-pager", "poster", "vision", "paper", "template", "concept", "note", "proposal", "note", "implementation", "plan", "simple", "table", "venture", "pack", "letter", "courrier", "meeting", "report", "compte", "rendu", "brochure", "law", "page", "counterpart", "index", "front", "page", "copy", "one", "and", "fill", "it" ] }, { "id": "resource-a9d437fede", "title": "The user manual", "domain": "UI", "domains": [ "UI" ], "kind": "Workflow recipe", "summary": "The software at its best: real screens, plain words, made again by script.", "href": "manuals/index.html", "sourcePath": "manuals/index.html", "thumbnail": "assets/thumbs/cards/card-user-manual.png", "keywords": [ "manuals/index.html", "manual", "mode", "emploi", "captures", "software", "journeys", "steps", "user", "manual", "manuel", "utilisateur", "mode", "d'emploi", "documentation", "software", "screens", "screen", "captures", "municipality", "mairie", "journeys", "steps", "how", "to", "use", "the", "app" ] }, { "id": "resource-438ce1e8d6", "title": "The UI review", "domain": "UI", "domains": [ "UI" ], "kind": "Review method", "summary": "Real screens judged, findings with repairs in plain words.", "href": "ui-review/index.html", "sourcePath": "ui-review/index.html", "thumbnail": "assets/thumbs/cards/card-ui-review.png", "keywords": [ "ui-review/index.html", "review", "findings", "repairs", "critique", "verdict", "page-review", "audit", "grayscale", "ui", "review", "revue", "ecran", "screen", "screens", "findings", "repairs", "kennedy", "teletubbies", "critique", "verdict", "reviewer", "page-review", "review", "this", "page", "review", "my", "screens", "audit", "my", "page", "comments", "how", "to", "improve", "token", "audit", "composition", "audit", "grayscale", "measure", "line", "length", "phone", "check", "document", "or", "piece", "mode", "gate" ] }, { "id": "resource-358cd08484", "title": "The design benches", "domain": "References", "domains": [ "References", "Images", "UI" ], "kind": "Example collection", "summary": "Each project's visual ideas before they are decided.", "href": "benches/index.html", "sourcePath": "benches/index.html", "thumbnail": "assets/thumbs/benches/abc-colombia.jpg", "keywords": [ "benches/index.html", "bench", "banco", "de", "diseno", "parked", "palettes", "layouts", "refused", "bench", "benches", "design", "bench", "banco", "de", "diseno", "project", "ideas", "parked", "palettes", "layouts", "photographs", "refused", "tried", "rejected", "central", "scrc", "abc", "colombia", "keep", "this", "store", "this", "for", "later", "reference" ] }, { "id": "resource-d880ab3088", "title": "The register of a document", "domain": "UI", "domains": [ "UI" ], "kind": "Writing standard", "summary": "Eight ways to open, one voice held to the end.", "href": "register/index.html", "sourcePath": "register/index.html", "thumbnail": "assets/thumbs/cards/card-register.jpg", "keywords": [ "register/index.html", "tone", "voice", "style", "narrative", "sounds", "like", "ai", "bluf", "scene", "metaphor", "register", "registro", "narrativo", "tone", "tone", "of", "voice", "style", "how", "the", "text", "sounds", "writing", "voice", "opening", "sentence", "sections", "concept", "note", "donor", "proposal", "speech", "ministerial", "note", "bluf", "scene", "metaphor", "sounds", "like", "ai", "rewrite", "the", "tone", "editorial", "voice", "consistency", "unslop" ] }, { "id": "resource-808956192e", "title": "Make your own speaker", "domain": "Speakers", "domains": [ "Speakers", "Voice" ], "kind": "Creation recipe", "summary": "Three roads: one that exists, a new one, or your own face.", "href": "people/make-your-own.html", "sourcePath": "people/make-your-own.html", "thumbnail": "assets/thumbs/cards/card-make-your-own.jpg", "keywords": [ "people/make-your-own.html", "wizard", "my", "own", "face", "colleague", "nelson", "create", "avatar", "make", "your", "own", "speaker", "avatar", "my", "own", "face", "my", "voice", "yourself", "colleague", "nelson", "wizard", "how", "do", "i", "get", "one", "create" ] }, { "id": "resource-930b47ef1b", "title": "Voice", "domain": "Voice", "domains": [ "Voice" ], "kind": "Capability collection", "summary": "A page that reads itself, and dictation into any window.", "href": "voice/index.html", "sourcePath": "voice/index.html", "thumbnail": "assets/thumbs/cards/card-voice.jpg", "keywords": [ "voice/index.html", "voice", "narration", "zalia", "grace", "clone", "pronunciation", "dictation", "freeflow", "wispr", "speak", "to", "the", "machine", "voice", "narration", "read", "aloud", "clone", "pronunciation", "saying", "names", "right", "dictation", "dictate", "speak", "type", "freeflow", "wispr", "flow", "fn", "key", "vocabulary", "slash", "commands", "transcription", "whisper", "groq", "speech", "to", "text", "speak", "to", "the", "machine", "two-way", "answer", "back", "realtime", "openai", "protocol", "sovereign", "vibevoice", "transcribe", "meeting", "who", "spoke", "when" ] }, { "id": "resource-54124ea461", "title": "The guide", "domain": "Voice", "domains": [ "Voice", "UI" ], "kind": "Interaction pattern", "summary": "Build a conversational guide and verify speech, interruption and recovery.", "href": "guide/index.html", "sourcePath": "guide/index.html", "thumbnail": "assets/thumbs/cards/card-guide.jpg", "keywords": [ "guide/index.html", "guide", "conversation", "dialogue", "interrupt", "rag", "grounded", "elevenlabs", "guide", "conversation", "talk", "dialogue", "interrupt", "answers", "questions", "rag", "grounded", "zalia", "valentina", "grace", "elevenlabs" ] }, { "id": "resource-486e92bf92", "title": "Motion", "domain": "UI", "domains": [ "UI" ], "kind": "Technique collection", "summary": "The map of the family: the one secret, the four moments, the shelf.", "href": "motion/index.html", "sourcePath": "motion/index.html", "thumbnail": "assets/thumbs/pages/resource-486e92bf92.jpg", "keywords": [ "motion/index.html", "motion", "front", "door", "map", "animation", "video", "motion", "animated", "site", "video", "scroll", "scrub", "entrance", "ambient", "loop", "melt", "parallax", "curtain", "recipes", "moving", "page" ] }, { "id": "resource-7a6278ccd4", "title": "The galleries", "domain": "References", "domains": [ "References", "Voice", "Images", "UI" ], "kind": "Reference collection", "summary": "Where to look first: sites, screens, motion, delightful details.", "href": "inspiration/index.html", "sourcePath": "inspiration/index.html", "thumbnail": "motion/assets/ais_0108.jpg", "keywords": [ "inspiration/index.html", "inspiration", "galleries", "mobbin", "godly", "refero", "logos", "illustration", "icons", "shots", "inspiration", "liked", "models", "bank", "picked", "usual", "presentation", "10", "thumbnails", "modern-ui", "reference", "galleries", "pinterest", "godly", "minimal", "gallery", "mobbin", "sceneai", "motionsites", "trends", "decoded", "builds", "designspells", "design", "spells", "delight", "delightful", "detail", "moment", "micro-interaction", "fun", "logo", "logos", "logotype", "mark", "symbol", "trademark", "wordmark", "logobook", "logoarchive", "logggos", "logoinspo", "logosystem", "illustration", "illustrations", "drawing", "spot", "graphic", "drawkit", "blush", "absurd", "sketchvalley", "icons8", "ouch", "kinetics", "colorion", "spring", "physics", "interaction", "60fps", "motion", "library", "shots", "recorded", "interactions", "gesture", "bottom", "sheet", "card", "stack", "pull", "morph", "liquid", "glass", "swiftui", "breakdown", "design.md", "refero", "styles", "ready", "design", "system", "minimal", "gallery", "anime.js", "animation", "engine", "aceternity", "landing", "components", "vibeindex", "tool", "directory", "taste", "skill", "design", "dna", "extract", "design", "tokens", "from", "a", "site", "what", "makes", "this", "site", "good", "reverse", "engineer", "a", "design", "copy", "this", "site's", "style", "tastelab", "senlin", "celia", "space", "registry", "spaceregistry", "dot-matrix", "globe", "orange", "purple", "violet", "accent", "unoosa", "outer", "space", "objects", "cyprus" ] }, { "id": "resource-256960b695", "title": "Taste", "domain": "References", "domains": [ "References", "Voice", "Images", "UI" ], "kind": "Principle", "summary": "Frank's verdicts on how things should look.", "href": "taste.html", "sourcePath": "taste.html", "thumbnail": "assets/thumbs/cards/card-taste.jpg", "keywords": [ "taste.html", "taste", "verdicts", "refused", "approved", "art", "director", "look", "taste", "verdict", "verdicts", "refused", "refusal", "rejected", "what", "frank", "refused", "what", "he", "does", "not", "want", "judgement", "criteria", "look", "and", "feel", "dark", "theme", "three", "options", "baseline", "ux", "nine", "rules", "skeleton", "beautiful", "ugly", "colour", "color" ] }, { "id": "resource-e9e67919cc", "title": "The speaker factory", "domain": "Speakers", "domains": [ "Speakers", "Voice" ], "kind": "Workflow", "summary": "Compare production routes, with their inputs, limits and evidence.", "href": "people/factory.html", "sourcePath": "people/factory.html", "thumbnail": "assets/thumbs/factory-page.png", "keywords": [ "people/factory.html", "factory", "cost", "gpu", "runpod", "magic", "hour", "master", "clip", "mouth", "factory", "speaker", "factory", "build", "your", "own", "cost", "centimes", "cheap", "rented", "computer", "gpu", "runpod", "magic", "hour", "master", "clip", "mouth", "video", "born", "native" ] }, { "id": "resource-9b080aff02", "title": "Send mail as your domain", "domain": "Workflows", "domains": [ "Workflows", "UI" ], "kind": "Publishing recipe", "summary": "Free, no server, replies included.", "href": "mail/branded-mail.html", "sourcePath": "mail/branded-mail.html", "thumbnail": "assets/thumbs/cards/card-mail2.jpg", "keywords": [ "mail/branded-mail.html", "mail", "email", "domain", "sender", "brevo", "dns", "invitations", "mail", "email", "domain", "sender", "brevo", "dns", "free", "invitations", "send", "a", "message", "reaches", "someone" ] }, { "id": "resource-94ed68a0df", "title": "Show a body, and put it straight", "domain": "Images", "domains": [ "Images", "Video" ], "kind": "Explanatory technique", "summary": "A foot collapsed, an axis bent, then put right in the same picture.", "href": "body/index.html", "sourcePath": "body/index.html", "thumbnail": "assets/thumbs/cards/card-body.jpg", "keywords": [ "body/index.html", "body", "anatomy", "foot", "spine", "posture", "before", "after", "correction", "x-ray", "skeleton", "motion", "body", "anatomy", "foot", "spine", "column", "posture", "straight", "crooked", "bent", "deformed", "before", "after", "before", "and", "after", "correction", "fix", "insole", "medical", "x-ray", "ghost", "body", "skeleton", "bones", "plumb", "line", "axis", "morph", "blender", "z-anatomy", "proof", "advert", "demonstration", "motion", "film", "drawing" ] }, { "id": "resource-8ec9a00bfd", "title": "The studio spine", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "What the studio is and where each thing lives, for a session reading files.", "href": "catalogue/read/resource-8ec9a00bfd.html", "sourcePath": "README.md", "thumbnail": "assets/thumbs/pages/resource-8ec9a00bfd.jpg", "keywords": [ "README.md", "readme", "start", "here", "routing" ] }, { "id": "resource-ea2bfff860", "title": "Bench: Sistema ABC", "domain": "References", "domains": [ "References" ], "kind": "Recipe", "summary": "Palettes, references and parked ideas for the Colombia system.", "href": "benches/abc-colombia/index.html", "sourcePath": "benches/abc-colombia/index.html", "thumbnail": "benches/abc-colombia/design-bench-refs/palette-tangent-thumb.jpg", "keywords": [ "benches/abc-colombia/index.html", "abc", "colombia", "bench", "banco", "minambiente" ] }, { "id": "resource-41bbf9bcc6", "title": "Bench: Central", "domain": "References", "domains": [ "References" ], "kind": "Recipe", "summary": "Palettes, references and parked ideas for Central.", "href": "benches/central/index.html", "sourcePath": "benches/central/index.html", "thumbnail": "benches/central/design-bench-refs/anillo-capacidades-thumb.jpg", "keywords": [ "benches/central/index.html", "central", "bench", "design", "ideas" ] }, { "id": "resource-5258cea9c7", "title": "Colour says which world a page belongs to", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Which paper and which colour a page wears, by family.", "href": "catalogue/read/resource-5258cea9c7.html", "sourcePath": "document-families.md", "thumbnail": "assets/thumbs/pages/resource-5258cea9c7.jpg", "keywords": [ "document-families.md", "colour", "color", "paper", "family", "beige", "brand" ] }, { "id": "resource-8ebe49057b", "title": "Documentation design", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "The sourced digest under the rules: twenty reading patterns.", "href": "catalogue/read/resource-8ebe49057b.html", "sourcePath": "documentation-design.md", "thumbnail": "assets/thumbs/pages/resource-8ebe49057b.jpg", "keywords": [ "documentation-design.md", "digest", "why", "sources", "reading", "patterns", "measure" ] }, { "id": "resource-758b2bc968", "title": "documentation-kit.css", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "The paste-from stylesheet: every block, on house tokens.", "href": "catalogue/read/resource-758b2bc968.html", "sourcePath": "documentation-kit.css", "thumbnail": "assets/thumbs/pages/resource-758b2bc968.jpg", "keywords": [ "documentation-kit.css", "css", "code", "segment", "callout", "steps", "stat", "tokens" ] }, { "id": "resource-9696d074c6", "title": "The animation studio, annex", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "External material kept whole, behind the film recipes.", "href": "catalogue/read/resource-9696d074c6.html", "sourcePath": "film-annex-animation-studio.md", "thumbnail": "assets/thumbs/pages/resource-9696d074c6.jpg", "keywords": [ "film-annex-animation-studio.md", "annex", "animation", "studio", "reference", "film" ] }, { "id": "resource-07232faf41", "title": "Concept note, the format", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The opening paper of any development, with its drawing.", "href": "catalogue/read/resource-07232faf41.html", "sourcePath": "formats/concept-note/concept-note.md", "thumbnail": "assets/thumbs/pages/resource-07232faf41.jpg", "keywords": [ "formats/concept-note/concept-note.md", "concept", "note", "format", "drawing", "standing", "procedure" ] }, { "id": "resource-297dcec9ac", "title": "Implementation plan, the format", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "Screens as the contract, phases with gates.", "href": "catalogue/read/resource-297dcec9ac.html", "sourcePath": "formats/implementation-plan/implementation-plan.md", "thumbnail": "assets/thumbs/pages/resource-297dcec9ac.jpg", "keywords": [ "formats/implementation-plan/implementation-plan.md", "implementation", "plan", "change", "proposal", "eregistrations", "format" ] }, { "id": "resource-7bb9ba9592", "title": "Itinerary card, a worked layout", "domain": "UI", "domains": [ "UI" ], "kind": "Example", "summary": "A layout example; its dates and locations are not current advice.", "href": "formats/itinerary-card/exemplar.html", "sourcePath": "formats/itinerary-card/exemplar.html", "thumbnail": "assets/thumbs/pages/resource-7bb9ba9592.jpg", "keywords": [] }, { "id": "resource-32db8b1683", "title": "Proposal note, a real one", "domain": "UI", "domains": [ "UI" ], "kind": "Recipe", "summary": "A decision carried on one page: problem, options, recommendation.", "href": "formats/proposal-note/exemplar.html", "sourcePath": "formats/proposal-note/exemplar.html", "thumbnail": "assets/thumbs/pages/resource-32db8b1683.jpg", "keywords": [ "formats/proposal-note/exemplar.html", "proposal", "note", "concept", "guides", "law", "citizens" ] }, { "id": "resource-64e0dc0ebb", "title": "Proposal note, the format", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "What the note holds and how it closes.", "href": "catalogue/read/resource-64e0dc0ebb.html", "sourcePath": "formats/proposal-note/proposal-note.md", "thumbnail": "assets/thumbs/pages/resource-64e0dc0ebb.jpg", "keywords": [ "formats/proposal-note/proposal-note.md", "proposal", "note", "format", "definition" ] }, { "id": "resource-8c9e8cd790", "title": "Simple table, a real one", "domain": "UI", "domains": [ "UI" ], "kind": "Recipe", "summary": "One screen answering how it went, per item.", "href": "formats/simple-table/exemplar.html", "sourcePath": "formats/simple-table/exemplar.html", "thumbnail": "assets/thumbs/pages/resource-8c9e8cd790.jpg", "keywords": [ "formats/simple-table/exemplar.html", "table", "coverage", "verdict", "board", "badges", "stats" ] }, { "id": "resource-1d200c8b74", "title": "Simple table, the format", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The parts of the verdict board.", "href": "catalogue/read/resource-1d200c8b74.html", "sourcePath": "formats/simple-table/simple-table.md", "thumbnail": "assets/thumbs/pages/resource-1d200c8b74.jpg", "keywords": [ "formats/simple-table/simple-table.md", "table", "format", "definition", "coverage" ] }, { "id": "resource-f15b82fd93", "title": "Vision paper, the format", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "What the shape is and when it is right.", "href": "catalogue/read/resource-f15b82fd93.html", "sourcePath": "formats/vision-paper/format.md", "thumbnail": "assets/thumbs/pages/resource-f15b82fd93.jpg", "keywords": [ "formats/vision-paper/format.md", "vision", "paper", "format", "definition" ] }, { "id": "resource-f1ac8718c9", "title": "Vision paper, the page", "domain": "UI", "domains": [ "UI" ], "kind": "Recipe", "summary": "One argument, chapter by chapter, each opening with a summary.", "href": "formats/vision-paper/template.html", "sourcePath": "formats/vision-paper/template.html", "thumbnail": "assets/thumbs/pages/resource-f1ac8718c9.jpg", "keywords": [ "formats/vision-paper/template.html", "vision", "paper", "manifesto", "long-form", "thesis", "template" ] }, { "id": "resource-498b65ab2b", "title": "What the private sector does", "domain": "Principles", "domains": [ "Principles" ], "kind": "Recipe", "summary": "What companies such as Stripe do with forms today, read against the public-sector practice on file.", "href": "forms-private-sector.html", "sourcePath": "forms-private-sector.html", "thumbnail": "assets/thumbs/pages/resource-498b65ab2b.jpg", "keywords": [] }, { "id": "resource-94edd103a1", "title": "The state of the art in form design", "domain": "Principles", "domains": [ "Principles" ], "kind": "Recipe", "summary": "Form practices from primary sources, ordered by weight for a long expert filing done once or twice a year.", "href": "forms-state-of-the-art.html", "sourcePath": "forms-state-of-the-art.html", "thumbnail": "assets/thumbs/pages/resource-94edd103a1.jpg", "keywords": [] }, { "id": "resource-182c9d5ddd", "title": "Frank's baseline UX", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "The nine rules a page passes before Frank sees it.", "href": "catalogue/read/resource-182c9d5ddd.html", "sourcePath": "franks-baseline-ux.md", "thumbnail": "assets/thumbs/pages/resource-182c9d5ddd.jpg", "keywords": [ "franks-baseline-ux.md", "baseline", "nine", "rules", "pre-flight", "check" ] }, { "id": "resource-f0a4c6705c", "title": "Hosted voice agents, the state of the art", "domain": "Voice", "domains": [ "Voice" ], "kind": "Reference", "summary": "What you can buy today, judged on interruption.", "href": "catalogue/read/resource-f0a4c6705c.html", "sourcePath": "guide/research/hosted-platforms-06-08-2026.md", "thumbnail": "assets/thumbs/pages/resource-f0a4c6705c.jpg", "keywords": [ "guide/research/hosted-platforms-06-08-2026.md", "hosted", "platforms", "elevenlabs", "convai", "research", "barge-in" ] }, { "id": "resource-d7600334f9", "title": "Open voice agents, the state of the art", "domain": "Voice", "domains": [ "Voice" ], "kind": "Reference", "summary": "What you can host yourself, judged on interruption.", "href": "catalogue/read/resource-d7600334f9.html", "sourcePath": "guide/research/open-stacks-06-08-2026.md", "thumbnail": "assets/thumbs/pages/resource-d7600334f9.jpg", "keywords": [ "guide/research/open-stacks-06-08-2026.md", "open", "self-hosted", "stacks", "research", "barge-in", "sovereign" ] }, { "id": "resource-94e7f7c9df", "title": "Image models for skills", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Which image engine to pay for when a skill must repeat itself.", "href": "catalogue/read/resource-94e7f7c9df.html", "sourcePath": "image-models-for-skills-2026-08.md", "thumbnail": "assets/thumbs/pages/resource-94e7f7c9df.jpg", "keywords": [ "image-models-for-skills-2026-08.md", "imagen", "modelos", "image", "models", "engines", "cost", "skills", "nano", "banana" ] }, { "id": "resource-0f5efa1174", "title": "Impeccable integration", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Focus, states and the small details that finish a screen.", "href": "catalogue/read/resource-0f5efa1174.html", "sourcePath": "impeccable-integration.md", "thumbnail": "assets/thumbs/pages/resource-0f5efa1174.jpg", "keywords": [ "impeccable-integration.md", "focus", "states", "details", "vocabulary" ] }, { "id": "resource-a47052350d", "title": "Implementation proof pipeline", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "How a plan shows delivery and verification in one place.", "href": "catalogue/read/resource-a47052350d.html", "sourcePath": "implementation-proof-pipeline.md", "thumbnail": "assets/thumbs/pages/resource-a47052350d.jpg", "keywords": [ "implementation-proof-pipeline.md", "proof", "verification", "plan", "dates", "evidence" ] }, { "id": "resource-db05fd3ab0", "title": "The inbox", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Raw material parked, waiting to become a rule.", "href": "catalogue/read/resource-db05fd3ab0.html", "sourcePath": "inbox.md", "thumbnail": "assets/thumbs/pages/resource-db05fd3ab0.jpg", "keywords": [ "inbox.md", "parked", "raw", "material", "extraction" ] }, { "id": "resource-70f7d8fc71", "title": "Model: Code Wiki", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Light without lights, 3D without models.", "href": "catalogue/read/resource-70f7d8fc71.html", "sourcePath": "inspiration/liked/codewiki-google/model.md", "thumbnail": "assets/thumbs/pages/resource-70f7d8fc71.jpg", "keywords": [ "inspiration/liked/codewiki-google/model.md", "codewiki", "google", "liked", "model", "captured" ] }, { "id": "resource-f4edaece86", "title": "The liked models", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "The sites Frank picked, captured and described.", "href": "inspiration/liked/index.html", "sourcePath": "inspiration/liked/index.html", "thumbnail": "inspiration/liked/interfaces-dev/thumb.jpg", "keywords": [ "inspiration/liked/index.html", "liked", "bank", "models", "picked", "chosen", "reference" ] }, { "id": "resource-6a3b4bd342", "title": "Model: interfaces.dev", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "A page that measures itself in front of you.", "href": "catalogue/read/resource-6a3b4bd342.html", "sourcePath": "inspiration/liked/interfaces-dev/model.md", "thumbnail": "assets/thumbs/pages/resource-6a3b4bd342.jpg", "keywords": [ "inspiration/liked/interfaces-dev/model.md", "interfaces", "dev", "liked", "model", "magazine" ] }, { "id": "resource-6e66d15bcf", "title": "Model: L'étape d'après", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "A photograph as the room, one italic word as the accent.", "href": "catalogue/read/resource-6e66d15bcf.html", "sourcePath": "inspiration/liked/letape-dapres/model.md", "thumbnail": "assets/thumbs/pages/resource-6e66d15bcf.jpg", "keywords": [ "inspiration/liked/letape-dapres/model.md", "letape", "dapres", "liked", "model", "photograph" ] }, { "id": "resource-bf1bb22b36", "title": "Shown log", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Every site already brought to Frank, so none is shown twice.", "href": "catalogue/read/resource-bf1bb22b36.html", "sourcePath": "inspiration/liked/shown-log.md", "thumbnail": "assets/thumbs/pages/resource-bf1bb22b36.jpg", "keywords": [ "inspiration/liked/shown-log.md", "shown", "log", "rounds", "thumbnails", "never", "repeat" ] }, { "id": "resource-aa912d7ad9", "title": "Model: Stripe Press", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Books as objects, on a quiet page.", "href": "catalogue/read/resource-aa912d7ad9.html", "sourcePath": "inspiration/liked/stripe-press/model.md", "thumbnail": "assets/thumbs/pages/resource-aa912d7ad9.jpg", "keywords": [ "inspiration/liked/stripe-press/model.md", "stripe", "press", "liked", "model", "books" ] }, { "id": "resource-c2d5b82b8f", "title": "Where to find UI ideas", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "The classified list: what each gallery is best for.", "href": "catalogue/read/resource-c2d5b82b8f.html", "sourcePath": "inspiration/references.md", "thumbnail": "assets/thumbs/pages/resource-c2d5b82b8f.jpg", "keywords": [ "inspiration/references.md", "references", "screensdesign", "mobbin", "layers", "classification" ] }, { "id": "resource-8ff3464713", "title": "ScreensDesign, studied", "domain": "References", "domains": [ "References" ], "kind": "Recipe", "summary": "What the site holds and how to use it without copying.", "href": "inspiration/screensdesign.html", "sourcePath": "inspiration/screensdesign.html", "thumbnail": "inspiration/voxpopuli-app-arrival-current.png", "keywords": [ "inspiration/screensdesign.html", "screensdesign", "studied", "reference", "app", "screens" ] }, { "id": "resource-ce8c90315d", "title": "ScreensDesign, studied", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "The source behind inspiration/screensdesign.html.", "href": "catalogue/read/resource-ce8c90315d.html", "sourcePath": "inspiration/screensdesign.md", "thumbnail": "assets/thumbs/cards/card-formats.jpg", "keywords": [ "inspiration/screensdesign.md", "inspiration/screensdesign.md", "inspiration/screensdesign.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-027ff8200c", "title": "Web design trends 2026", "domain": "References", "domains": [ "References" ], "kind": "Recipe", "summary": "About twenty-five trends, tagged, each marked for a readable page.", "href": "inspiration/trends.html", "sourcePath": "inspiration/trends.html", "thumbnail": "assets/thumbs/pages/resource-027ff8200c.jpg", "keywords": [ "inspiration/trends.html", "trends", "2026", "3d", "animation", "typography", "ai", "bento", "minimalism" ] }, { "id": "resource-400fe2de52", "title": "Digest: the two-plate build", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Building a site from two plates, start to finish.", "href": "catalogue/read/resource-400fe2de52.html", "sourcePath": "inspiration/viktoroddy/5ZAsjUoMx1Y-two-plate-build.md", "thumbnail": "assets/thumbs/pages/resource-400fe2de52.jpg", "keywords": [ "inspiration/viktoroddy/5ZAsjUoMx1Y-two-plate-build.md", "viktoroddy", "fable", "two", "plate", "build", "tutorial" ] }, { "id": "resource-34cdb6f9e6", "title": "Digest: the head-turn hero", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "A 3D head that follows the scroll.", "href": "catalogue/read/resource-34cdb6f9e6.html", "sourcePath": "inspiration/viktoroddy/9RiV-_MUiCI-head-turn-hero.md", "thumbnail": "assets/thumbs/pages/resource-34cdb6f9e6.jpg", "keywords": [ "inspiration/viktoroddy/9RiV-_MUiCI-head-turn-hero.md", "viktoroddy", "3d", "animation", "head", "turn", "hero" ] }, { "id": "resource-3d7efaf818", "title": "Digest: reskinning a 3D site", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Taking one built site and giving it a new face.", "href": "catalogue/read/resource-3d7efaf818.html", "sourcePath": "inspiration/viktoroddy/IeR5ZMKssSc-sonnet5-reskin-3d-sites.md", "thumbnail": "assets/thumbs/pages/resource-3d7efaf818.jpg", "keywords": [ "inspiration/viktoroddy/IeR5ZMKssSc-sonnet5-reskin-3d-sites.md", "viktoroddy", "sonnet", "reskin", "3d", "sites" ] }, { "id": "resource-b5c4d4fd84", "title": "Digest: GPT Sol for web design", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "What the model changes for a page builder.", "href": "catalogue/read/resource-b5c4d4fd84.html", "sourcePath": "inspiration/viktoroddy/YhF8Qg2GL80-gpt-sol-web-design.md", "thumbnail": "assets/thumbs/pages/resource-b5c4d4fd84.jpg", "keywords": [ "inspiration/viktoroddy/YhF8Qg2GL80-gpt-sol-web-design.md", "viktoroddy", "gpt", "sol", "web", "design" ] }, { "id": "resource-71cb9ae7b1", "title": "Digest: Fable for web design", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "The workflow, from prompt to published page.", "href": "catalogue/read/resource-71cb9ae7b1.html", "sourcePath": "inspiration/viktoroddy/_JF_s-ZRTyY-fable5-web-design.md", "thumbnail": "assets/thumbs/pages/resource-71cb9ae7b1.jpg", "keywords": [ "inspiration/viktoroddy/_JF_s-ZRTyY-fable5-web-design.md", "viktoroddy", "fable", "web", "design", "workflow" ] }, { "id": "resource-028ecb5951", "title": "Digest: motion transfer", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Moving a motion from one page to another.", "href": "catalogue/read/resource-028ecb5951.html", "sourcePath": "inspiration/viktoroddy/bhietcNpXw8-motion-transfer-build.md", "thumbnail": "assets/thumbs/pages/resource-028ecb5951.jpg", "keywords": [ "inspiration/viktoroddy/bhietcNpXw8-motion-transfer-build.md", "viktoroddy", "motion", "transfer", "autopilot" ] }, { "id": "resource-f23f53d02f", "title": "Digest: the scroll website", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "A page built around one long scroll.", "href": "catalogue/read/resource-f23f53d02f.html", "sourcePath": "inspiration/viktoroddy/kM7cHTf-RAg-grok-scroll-website.md", "thumbnail": "assets/thumbs/pages/resource-f23f53d02f.jpg", "keywords": [ "inspiration/viktoroddy/kM7cHTf-RAg-grok-scroll-website.md", "viktoroddy", "grok", "scroll", "website" ] }, { "id": "resource-e13be21a06", "title": "Digest: the parallax image split", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Pages that feel like films.", "href": "catalogue/read/resource-e13be21a06.html", "sourcePath": "inspiration/viktoroddy/kf3OpxxmC9E-parallax-image-split.md", "thumbnail": "assets/thumbs/pages/resource-e13be21a06.jpg", "keywords": [ "inspiration/viktoroddy/kf3OpxxmC9E-parallax-image-split.md", "viktoroddy", "parallax", "image", "split", "movies" ] }, { "id": "resource-968a24da74", "title": "Digest: Kimi for web design", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Another engine, the same test.", "href": "catalogue/read/resource-968a24da74.html", "sourcePath": "inspiration/viktoroddy/sUeTik8Sf4o-kimi-k3-web-designs.md", "thumbnail": "assets/thumbs/pages/resource-968a24da74.jpg", "keywords": [ "inspiration/viktoroddy/sUeTik8Sf4o-kimi-k3-web-designs.md", "viktoroddy", "kimi", "web", "designs" ] }, { "id": "resource-e2bb02d7a6", "title": "Digest: animation with Figma AI", "domain": "References", "domains": [ "References" ], "kind": "Reference", "summary": "Where Figma fits once the AI does the moving.", "href": "catalogue/read/resource-e2bb02d7a6.html", "sourcePath": "inspiration/viktoroddy/uEJIbPQYq18-figma-ai-animation.md", "thumbnail": "assets/thumbs/pages/resource-e2bb02d7a6.jpg", "keywords": [ "inspiration/viktoroddy/uEJIbPQYq18-figma-ai-animation.md", "viktoroddy", "figma", "ai", "animation", "tutorial" ] }, { "id": "resource-afaf7b0ad7", "title": "The Kennedy shelf", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Erik Kennedy's free material, digested. Start here.", "href": "catalogue/read/resource-afaf7b0ad7.html", "sourcePath": "kennedy/_index.md", "thumbnail": "assets/thumbs/pages/resource-afaf7b0ad7.jpg", "keywords": [ "kennedy/_index.md", "learnui", "erik", "kennedy", "shelf", "corpus" ] }, { "id": "resource-f40ac3d1ed", "title": "Against systems", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Kennedy's case against our token machine.", "href": "catalogue/read/resource-f40ac3d1ed.html", "sourcePath": "kennedy/against-systems.md", "thumbnail": "assets/thumbs/pages/resource-f40ac3d1ed.jpg", "keywords": [ "kennedy/against-systems.md", "kennedy", "grids", "type", "scales", "color", "theory", "beginners" ] }, { "id": "resource-1882e03a9b", "title": "Colour, one base varied by recipe", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "One colour, many uses, no palette generator.", "href": "catalogue/read/resource-1882e03a9b.html", "sourcePath": "kennedy/color.md", "thumbnail": "assets/thumbs/pages/resource-1882e03a9b.jpg", "keywords": [ "kennedy/color.md", "kennedy", "color", "colour", "palette", "recipe" ] }, { "id": "resource-a17b8b7ba8", "title": "Hierarchy, spacing, light and alignment", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Rank by taking away, not by adding.", "href": "catalogue/read/resource-a17b8b7ba8.html", "sourcePath": "kennedy/hierarchy-and-spacing.md", "thumbnail": "assets/thumbs/pages/resource-a17b8b7ba8.jpg", "keywords": [ "kennedy/hierarchy-and-spacing.md", "kennedy", "de-emphasis", "spacing", "light", "alignment" ] }, { "id": "resource-dc0d29e742", "title": "Intuitive UX", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Locality, controls, and the squint test.", "href": "catalogue/read/resource-dc0d29e742.html", "sourcePath": "kennedy/intuitive-ux.md", "thumbnail": "assets/thumbs/pages/resource-dc0d29e742.jpg", "keywords": [ "kennedy/intuitive-ux.md", "kennedy", "intuitive", "squint", "controls", "locality" ] }, { "id": "resource-8a65f74812", "title": "Practice methods", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "How Kennedy trains the eye.", "href": "catalogue/read/resource-8a65f74812.html", "sourcePath": "kennedy/practice-methods.md", "thumbnail": "assets/thumbs/pages/resource-8a65f74812.jpg", "keywords": [ "kennedy/practice-methods.md", "kennedy", "practice", "exercises", "training" ] }, { "id": "resource-cb53a639d7", "title": "Kennedy's free tools", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "What each one is for, and when to open it.", "href": "catalogue/read/resource-cb53a639d7.html", "sourcePath": "kennedy/tools.md", "thumbnail": "assets/thumbs/pages/resource-cb53a639d7.jpg", "keywords": [ "kennedy/tools.md", "kennedy", "tools", "free", "browser" ] }, { "id": "resource-056b9dd0db", "title": "Typography and fonts", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "Sizes, pairing, and the safe free fonts.", "href": "catalogue/read/resource-056b9dd0db.html", "sourcePath": "kennedy/typography-and-fonts.md", "thumbnail": "assets/thumbs/pages/resource-056b9dd0db.jpg", "keywords": [ "kennedy/typography-and-fonts.md", "kennedy", "type", "font", "size", "pairing", "google", "fonts" ] }, { "id": "resource-a40c88b2c8", "title": "The paper change", "domain": "UI", "domains": [ "UI" ], "kind": "Recipe", "summary": "Two paper colours had crept into the studio; the sweep of 29-08-2026 made them one.", "href": "lookup/the-paper-change.html", "sourcePath": "lookup/the-paper-change.html", "thumbnail": "assets/thumbs/pages/resource-a40c88b2c8.jpg", "keywords": [] }, { "id": "resource-1b84fe7b50", "title": "Give a site its own mail", "domain": "Workflows", "domains": [ "Workflows" ], "kind": "Reference", "summary": "The source behind mail/branded-mail.html.", "href": "catalogue/read/resource-1b84fe7b50.html", "sourcePath": "mail/branded-mail.md", "thumbnail": "assets/thumbs/pages/resource-1b84fe7b50.jpg", "keywords": [ "mail/branded-mail.md", "mail/branded-mail.md", "mail/branded-mail.html", "source", "markdown", "machine", "file" ] }, { "id": "resource-17fc09f326", "title": "Images and page entrances, the audit", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "What we shipped, and the minimum standard agreed.", "href": "catalogue/read/resource-17fc09f326.html", "sourcePath": "motion/audit-images-and-entrances-08-08-2026.md", "thumbnail": "assets/thumbs/pages/resource-17fc09f326.jpg", "keywords": [ "motion/audit-images-and-entrances-08-08-2026.md", "audit", "images", "entrances", "flat", "coded", "drawings", "standard" ] }, { "id": "resource-27cada4122", "title": "Frank's criteria for motion", "domain": "UI", "domains": [ "UI" ], "kind": "Recipe", "summary": "The numbered verdicts on what may move, and how much.", "href": "motion/criteria.html", "sourcePath": "motion/criteria.html", "thumbnail": "assets/thumbs/pages/resource-27cada4122.jpg", "keywords": [ "motion/criteria.html", "criteria", "verdicts", "rules", "motion", "judgment" ] }, { "id": "resource-b6c10e0a79", "title": "Frank's criteria for motion", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind motion/criteria.html.", "href": "catalogue/read/resource-b6c10e0a79.html", "sourcePath": "motion/criteria.md", "thumbnail": "assets/thumbs/cards/card-film.jpg", "keywords": [ "motion/criteria.md", "motion/criteria.md", "motion/criteria.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-ad902a5fb0", "title": "Entrance recipes", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "What makes a first screen move beautifully.", "href": "catalogue/read/resource-ad902a5fb0.html", "sourcePath": "motion/entrance-recipes-from-viktoroddy-08-08-2026.md", "thumbnail": "assets/thumbs/pages/resource-ad902a5fb0.jpg", "keywords": [ "motion/entrance-recipes-from-viktoroddy-08-08-2026.md", "entrance", "recipes", "first", "screen", "hero" ] }, { "id": "resource-aa8f480b0a", "title": "Make pages move", "domain": "UI", "domains": [ "UI" ], "kind": "Recipe", "summary": "Video under the page, scroll, entrances: the recipes.", "href": "motion/motion-design.html", "sourcePath": "motion/motion-design.html", "thumbnail": "assets/thumbs/pages/resource-aa8f480b0a.jpg", "keywords": [ "motion/motion-design.html", "recipes", "scroll", "scrub", "entrance", "ambient", "loop", "parallax", "curtain" ] }, { "id": "resource-974103214b", "title": "Make pages move", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind motion/motion-design.html.", "href": "catalogue/read/resource-974103214b.html", "sourcePath": "motion/motion-design.md", "thumbnail": "assets/thumbs/cards/card-film.jpg", "keywords": [ "motion/motion-design.md", "motion/motion-design.md", "motion/motion-design.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-49b36d351d", "title": "The 3D web meta-prompt", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The prompt that makes a page look three-dimensional.", "href": "motion/reference-3d-web-prompt.html", "sourcePath": "motion/reference-3d-web-prompt.html", "thumbnail": "assets/thumbs/pages/resource-49b36d351d.jpg", "keywords": [ "motion/reference-3d-web-prompt.html", "3d", "web", "meta", "prompt", "three", "dimensional" ] }, { "id": "resource-21c3022b37", "title": "The 3D web meta-prompt", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind motion/reference-3d-web-prompt.html.", "href": "catalogue/read/resource-21c3022b37.html", "sourcePath": "motion/reference-3d-web-prompt.md", "thumbnail": "assets/thumbs/cards/card-film.jpg", "keywords": [ "motion/reference-3d-web-prompt.md", "motion/reference-3d-web-prompt.md", "motion/reference-3d-web-prompt.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-19c2f0c2e3", "title": "The Marcus prompt", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "A person standing on top of the site.", "href": "motion/reference-marcus-prompt.html", "sourcePath": "motion/reference-marcus-prompt.html", "thumbnail": "assets/thumbs/pages/resource-19c2f0c2e3.jpg", "keywords": [ "motion/reference-marcus-prompt.html", "marcus", "person", "on", "top", "prompt", "recreation" ] }, { "id": "resource-a4693fac9b", "title": "The Marcus prompt", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind motion/reference-marcus-prompt.html.", "href": "catalogue/read/resource-a4693fac9b.html", "sourcePath": "motion/reference-marcus-prompt.md", "thumbnail": "assets/thumbs/cards/card-film.jpg", "keywords": [ "motion/reference-marcus-prompt.md", "motion/reference-marcus-prompt.md", "motion/reference-marcus-prompt.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-252af3d714", "title": "motionsites.ai, decoded", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "How the gallery's pages are built.", "href": "motion/reference-motionsites-analysis.html", "sourcePath": "motion/reference-motionsites-analysis.html", "thumbnail": "assets/thumbs/pages/resource-252af3d714.jpg", "keywords": [ "motion/reference-motionsites-analysis.html", "motionsites", "analysis", "decoded", "gallery" ] }, { "id": "resource-aa16aa8237", "title": "motionsites.ai, decoded", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind motion/reference-motionsites-analysis.html.", "href": "catalogue/read/resource-aa16aa8237.html", "sourcePath": "motion/reference-motionsites-analysis.md", "thumbnail": "assets/thumbs/cards/card-film.jpg", "keywords": [ "motion/reference-motionsites-analysis.md", "motion/reference-motionsites-analysis.md", "motion/reference-motionsites-analysis.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-24a896c3ae", "title": "The NovaAI prompt", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The recreation prompt, kept as written.", "href": "motion/reference-novaai-prompt.html", "sourcePath": "motion/reference-novaai-prompt.html", "thumbnail": "assets/thumbs/pages/resource-24a896c3ae.jpg", "keywords": [ "motion/reference-novaai-prompt.html", "novaai", "recreation", "prompt", "viktor", "oddy" ] }, { "id": "resource-7b7ba36e12", "title": "The NovaAI prompt", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind motion/reference-novaai-prompt.html.", "href": "catalogue/read/resource-7b7ba36e12.html", "sourcePath": "motion/reference-novaai-prompt.md", "thumbnail": "assets/thumbs/cards/card-film.jpg", "keywords": [ "motion/reference-novaai-prompt.md", "motion/reference-novaai-prompt.md", "motion/reference-novaai-prompt.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-29ff773d28", "title": "Google AI Studio, the full process", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "Beginner to finished page, step by step.", "href": "motion/reference-viktoroddy-aistudio.html", "sourcePath": "motion/reference-viktoroddy-aistudio.html", "thumbnail": "motion/assets/ais_0005.jpg", "keywords": [ "motion/reference-viktoroddy-aistudio.html", "google", "ai", "studio", "viktor", "oddy", "process", "tutorial" ] }, { "id": "resource-4fc4e489da", "title": "Google AI Studio, the full process", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind motion/reference-viktoroddy-aistudio.html.", "href": "catalogue/read/resource-4fc4e489da.html", "sourcePath": "motion/reference-viktoroddy-aistudio.md", "thumbnail": "assets/thumbs/cards/card-film.jpg", "keywords": [ "motion/reference-viktoroddy-aistudio.md", "motion/reference-viktoroddy-aistudio.md", "motion/reference-viktoroddy-aistudio.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-e8cfcc07e9", "title": "Higgsfield and Claude, the tutorial", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The animated-site method, summarized with thumbnails.", "href": "motion/reference-viktoroddy-tutorial.html", "sourcePath": "motion/reference-viktoroddy-tutorial.html", "thumbnail": "motion/assets/tut_f_001.jpg", "keywords": [ "motion/reference-viktoroddy-tutorial.html", "higgsfield", "mcp", "viktor", "oddy", "tutorial", "images", "style" ] }, { "id": "resource-46ea388217", "title": "How award-winning animated sites are sold", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The business behind the pages, from the same source.", "href": "motion/reference-yt-award-winning-animated-sites.html", "sourcePath": "motion/reference-yt-award-winning-animated-sites.html", "thumbnail": "motion/assets/yt_0003.jpg", "keywords": [ "motion/reference-yt-award-winning-animated-sites.html", "award", "winning", "animated", "sites", "selling", "clients" ] }, { "id": "resource-2c8697ac82", "title": "The entrance image, research", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The fastest reliable way to make one, and what it costs.", "href": "catalogue/read/resource-2c8697ac82.html", "sourcePath": "motion/research-entrance-images-08-08-2026.md", "thumbnail": "assets/thumbs/pages/resource-2c8697ac82.jpg", "keywords": [ "motion/research-entrance-images-08-08-2026.md", "research", "entrance", "image", "realistic", "moving", "cost" ] }, { "id": "resource-317051140d", "title": "Speaker brief template", "domain": "Speakers", "domains": [ "Speakers" ], "kind": "Template", "summary": "A blank reusable brief; fill in the speaker and asset references before use.", "href": "people/brief-plantilla.html", "sourcePath": "people/brief-plantilla.html", "thumbnail": "assets/thumbs/pages/resource-317051140d.jpg", "keywords": [] }, { "id": "resource-091777c9c2", "title": "The speakers, set straight (in Spanish)", "domain": "Speakers", "domains": [ "Speakers" ], "kind": "Recipe", "summary": "Three independent reviews of 31-08-2026: what is decided, what it really costs, and three fixes before spending more.", "href": "people/doctrine-audit-31-08.html", "sourcePath": "people/doctrine-audit-31-08.html", "thumbnail": "assets/thumbs/pages/resource-091777c9c2.jpg", "keywords": [] }, { "id": "resource-a997510327", "title": "Which engine animates the face", "domain": "Speakers", "domains": [ "Speakers" ], "kind": "Recipe", "summary": "Our factory against HeyGen, the same words, judged by ear and eye.", "href": "people/engine-comparison/index.html", "sourcePath": "people/engine-comparison/index.html", "thumbnail": "assets/thumbs/pages/resource-a997510327.jpg", "keywords": [ "people/engine-comparison/index.html", "engine", "comparison", "heygen", "quality", "test", "motor" ] }, { "id": "resource-0617f01e99", "title": "How a speaker is made", "domain": "Speakers", "domains": [ "Speakers" ], "kind": "Recipe", "summary": "The one sequence every session follows, each step with its tool, its check and its cost.", "href": "people/how-a-speaker-is-made.html", "sourcePath": "people/how-a-speaker-is-made.html", "thumbnail": "assets/thumbs/pages/resource-0617f01e99.jpg", "keywords": [] }, { "id": "resource-9c526bbd34", "title": "Where the speakers stand on conversation", "domain": "Speakers", "domains": [ "Speakers" ], "kind": "Reference", "summary": "What a speaker can answer today, and what it cannot.", "href": "catalogue/read/resource-9c526bbd34.html", "sourcePath": "people/interaction-status.md", "thumbnail": "assets/thumbs/pages/resource-9c526bbd34.jpg", "keywords": [ "people/interaction-status.md", "interaction", "conversation", "status", "presenting", "answering" ] }, { "id": "resource-9b6729005b", "title": "Take a speaker that exists", "domain": "Speakers", "domains": [ "Speakers" ], "kind": "Recipe", "summary": "The shortest road: a speaker already made.", "href": "people/speaker-existing.html", "sourcePath": "people/speaker-existing.html", "thumbnail": "assets/thumbs/pages/resource-9b6729005b.jpg", "keywords": [ "people/speaker-existing.html", "existing", "speaker", "road", "one" ] }, { "id": "resource-09506348b6", "title": "Make a new speaker", "domain": "Speakers", "domains": [ "Speakers" ], "kind": "Recipe", "summary": "Invent a person who is nobody real.", "href": "people/speaker-new.html", "sourcePath": "people/speaker-new.html", "thumbnail": "assets/thumbs/pages/resource-09506348b6.jpg", "keywords": [ "people/speaker-new.html", "new", "invented", "speaker", "road", "two" ] }, { "id": "resource-853b710e6d", "title": "Make an avatar of yourself", "domain": "Speakers", "domains": [ "Speakers" ], "kind": "Recipe", "summary": "Your own face and your own voice.", "href": "people/speaker-yourself.html", "sourcePath": "people/speaker-yourself.html", "thumbnail": "assets/thumbs/pages/resource-853b710e6d.jpg", "keywords": [ "people/speaker-yourself.html", "yourself", "avatar", "own", "face", "road", "three" ] }, { "id": "resource-3a351a0d40", "title": "Install our skills", "domain": "Principles", "domains": [ "Principles" ], "kind": "Recipe", "summary": "One command takes any recipe into your own Claude.", "href": "skills.html", "sourcePath": "skills.html", "thumbnail": "assets/thumbs/sites/letape-dapres-fr.jpg", "keywords": [ "skills.html", "npx", "skills", "install", "claude", "command" ] }, { "id": "resource-6f0cd742c7", "title": "Publishing the studio", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "How this site itself goes online, and where it lives.", "href": "catalogue/read/resource-6f0cd742c7.html", "sourcePath": "studio-online.md", "thumbnail": "assets/thumbs/pages/resource-6f0cd742c7.jpg", "keywords": [ "studio-online.md", "studio", "online", "smartrules", "publish", "deploy", "site" ] }, { "id": "resource-da06deaecf", "title": "Taste", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "The source behind taste.html.", "href": "catalogue/read/resource-da06deaecf.html", "sourcePath": "taste.md", "thumbnail": "assets/thumbs/cards/card-design-rules.jpg", "keywords": [ "taste.md", "taste.md", "taste.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-4e3e8a3c2a", "title": "Studio tool register", "domain": "Principles", "domains": [ "Principles" ], "kind": "Reference", "summary": "The source behind tools.html.", "href": "catalogue/read/resource-4e3e8a3c2a.html", "sourcePath": "tools.md", "thumbnail": "assets/thumbs/pages/resource-4e3e8a3c2a.jpg", "keywords": [ "tools.md", "tools.md", "tools.html", "source", "markdown", "machine", "file" ] }, { "id": "resource-8d426be22e", "title": "The new shape of the studio", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The seven families, the cards that move, the order of work.", "href": "ui-review/studio-reshuffle-15-08-2026.html", "sourcePath": "ui-review/studio-reshuffle-15-08-2026.html", "thumbnail": "ui-review/studio-reshuffle-15-08-2026/thumbs/publishing.jpg", "keywords": [ "ui-review/studio-reshuffle-15-08-2026.html", "reshuffle", "restructuring", "families", "chips", "cards", "proposal" ] }, { "id": "resource-7405127bdf", "title": "The new shape of the studio", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind ui-review/studio-reshuffle-15-08-2026.html.", "href": "catalogue/read/resource-7405127bdf.html", "sourcePath": "ui-review/studio-reshuffle-15-08-2026.md", "thumbnail": "assets/thumbs/cards/card-forms.jpg", "keywords": [ "ui-review/studio-reshuffle-15-08-2026.md", "ui-review/studio-reshuffle-15-08-2026.md", "ui-review/studio-reshuffle-15-08-2026.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-9881c23472", "title": "The studio, reviewed against itself", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "Ten findings from a walk through our own pages, 15-08-2026.", "href": "ui-review/studio-review-15-08-2026.html", "sourcePath": "ui-review/studio-review-15-08-2026.html", "thumbnail": "ui-review/studio-review-15-08-2026/formats-404.jpg", "keywords": [ "ui-review/studio-review-15-08-2026.html", "review", "studio", "findings", "adversarial", "usability" ] }, { "id": "resource-ca773b8e4a", "title": "The studio, reviewed against itself", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The source behind ui-review/studio-review-15-08-2026.html.", "href": "catalogue/read/resource-ca773b8e4a.html", "sourcePath": "ui-review/studio-review-15-08-2026.md", "thumbnail": "assets/thumbs/cards/card-forms.jpg", "keywords": [ "ui-review/studio-review-15-08-2026.md", "ui-review/studio-review-15-08-2026.md", "ui-review/studio-review-15-08-2026.html", "source", "markdown", "machine", "file" ], "listed": false }, { "id": "resource-9d3f74866c", "title": "Long read, the format", "domain": "UI", "domains": [ "UI" ], "kind": "Reference", "summary": "The measured anatomy of a long read: the index in the margin, the column, listen and share.", "href": "catalogue/read/resource-9d3f74866c.html", "sourcePath": "formats/long-read/long-read.md", "thumbnail": "assets/thumbs/pages/resource-9d3f74866c.jpg", "keywords": [ "formats/long-read/long-read.md", "long", "read", "format", "article", "index", "sticky", "listen", "share", "research", "institutional", "economic", "study" ] }, { "id": "resource-0031c79485", "title": "Long read, the template to copy", "domain": "UI", "domains": [ "UI" ], "kind": "Template", "summary": "A working page in the long-read shape: the index builds itself, share copies the address, listen appears only with a real recording.", "href": "formats/long-read/template.html", "sourcePath": "formats/long-read/template.html", "thumbnail": "assets/thumbs/pages/resource-0031c79485.jpg", "keywords": [ "formats/long-read/template.html", "long", "read", "template", "article", "paper", "study", "report", "index", "toc" ] }, { "id": "resource-family-video", "title": "A family video from clips and photos", "domain": "Video", "domains": [ "Video" ], "kind": "Workflow recipe", "summary": "Photos and phone clips turned into a short edited film with titles and your choice of music, without editing by hand.", "href": "family-video.html", "sourcePath": "family-video.html", "thumbnail": "assets/thumbs/goals/family-video.jpg", "keywords": [ "family", "video", "montage", "holiday", "clips", "photos", "titles", "music", "remotion", "family-video", "film de famille", "video de familia" ] }, { "id": "resource-product-films", "title": "A product video", "domain": "Video", "domains": [ "Video" ], "kind": "Workflow recipe", "summary": "A product film or service explainer in motion graphics: a reference film to match, three storyboards, stills before anything moves, notes in camera words.", "href": "product-films.html", "sourcePath": "product-films.html", "thumbnail": "assets/thumbs/goals/product-films.jpg", "keywords": [ "product film", "promotional video", "promo", "launch video", "explainer", "motion graphics", "storyboard", "whatships", "hyperframes", "remotion", "21st", "vidéo promotionnelle", "video promocional" ] }, { "id": "tool-freeflow", "title": "FreeFlow", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Dictation application", "summary": "Mac dictation with transcription, cleanup and spoken text editing.", "href": "compare/tools/freeflow.html", "thumbnail": "catalogue/tool-captures/freeflow.png", "keywords": [ "Microphone speech", "Text in the active app", "Default for the existing dictation workflow; choose the backend deliberately." ], "thumbnailEvidence": "Official FreeFlow application icon. Open the project source for its demonstration." }, { "id": "tool-fluidvoice", "title": "FluidVoice", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Dictation application", "summary": "Local-first dictation with selectable speech engines and optional enhancement.", "href": "compare/tools/fluidvoice.html", "thumbnail": "catalogue/tool-captures/fluidvoice.png", "keywords": [ "Microphone speech", "Recorded audio", "Dictated text", "File transcripts", "Choose when local processing is the priority; compare your own vocabulary against FreeFlow." ], "thumbnailEvidence": "Project screenshot: file transcription from the official source, retrieved 5 September 2026. Not a comparative quality test." }, { "id": "tool-wispr-flow", "title": "Wispr Flow", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Dictation application", "summary": "Commercial dictation product already considered in the personal workflow.", "href": "compare/tools/wispr-flow.html", "thumbnail": "catalogue/tool-captures/wispr-flow.jpg", "keywords": [ "Speech", "Dictated text", "Reconsider only if a new trial fixes the recognition problems previously reported." ], "thumbnailEvidence": "Official product preview from the official source, retrieved 5 September 2026. Not a comparative quality test." }, { "id": "tool-whisper", "title": "Whisper", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech-recognition model", "summary": "Open speech-recognition model for transcription, translation and language identification.", "href": "compare/tools/whisper.html", "thumbnail": "catalogue/tool-captures/whisper.jpg", "keywords": [ "Audio file", "Transcript", "Timed segments", "Use as a general multilingual transcription engine or baseline." ], "thumbnailEvidence": "Architecture diagram from the official OpenAI Whisper GitHub README, retrieved 14 September 2026." }, { "id": "tool-groq-whisper", "title": "Groq speech-to-text", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Hosted inference service", "summary": "Hosted transcription endpoint used by existing Studio ingestion workflows.", "href": "compare/tools/groq-whisper.html", "thumbnail": "catalogue/tool-captures/groq-whisper.jpg", "keywords": [ "Audio file", "Transcript", "Use when cloud processing is acceptable and an existing API workflow is useful." ], "thumbnailEvidence": "Headless capture of the official Groq homepage, retrieved 14 September 2026. Groq speech-to-text is one of Groq's hosted API models." }, { "id": "tool-parakeet", "title": "Parakeet TDT", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech-recognition model", "summary": "NVIDIA speech-recognition model used through compatible apps or runtimes.", "href": "compare/tools/parakeet.html", "thumbnail": "catalogue/tool-captures/parakeet.jpg", "keywords": [ "Audio", "Transcript", "Use through FluidVoice or a supported runtime when the selected model covers the language." ], "thumbnailEvidence": "Official Hugging Face social-thumbnail image for the NVIDIA Parakeet TDT model card, retrieved 14 September 2026." }, { "id": "tool-vibeasr-cpp", "title": "VibeASR.cpp / BitNet", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "CPU inference runtime", "summary": "CPU runtime for the compressed VibeVoice-ASR-BitNet model.", "href": "compare/tools/vibeasr-cpp.html", "thumbnail": "catalogue/tool-captures/vibeasr-cpp.jpg", "keywords": [ "Recorded audio", "Transcript", "Use for the recorded local transcription route when a plain transcript is sufficient." ], "thumbnailEvidence": "Report-overview image from the official Microsoft VibeASR.cpp GitHub README, retrieved 14 September 2026." }, { "id": "tool-vibevoice-asr", "title": "VibeVoice-ASR", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech-recognition model", "summary": "Model with documented structured transcription, speaker and timing output.", "href": "compare/tools/vibevoice-asr.html", "thumbnail": "catalogue/tool-captures/vibevoice-asr.jpg", "keywords": [ "Long audio recording", "Structured transcript", "Speaker labels", "Timestamps", "Evaluate when who-spoke-when output matters." ], "thumbnailEvidence": "Diarization-error-rate (DER) benchmark chart from the official Microsoft VibeVoice GitHub README, retrieved 14 September 2026." }, { "id": "tool-vibevoice-streaming", "title": "VibeVoice-ASR-Streaming", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Streaming recognition model", "summary": "Recently released streaming branch of the VibeVoice family.", "href": "compare/tools/vibevoice-streaming.html", "thumbnail": "catalogue/tool-captures/vibevoice-streaming.jpg", "keywords": [ "Incoming audio stream", "Incremental transcript", "Evaluate for live transcription where streaming is necessary." ], "thumbnailEvidence": "tcpWER benchmark chart from the official Microsoft VibeVoice GitHub README, retrieved 14 September 2026 (a different chart than the one used for VibeVoice-ASR)." }, { "id": "tool-elevenlabs", "title": "ElevenLabs", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Hosted voice platform", "summary": "Hosted speech generation, voice design, cloning and other audio capabilities.", "href": "compare/tools/elevenlabs.html", "thumbnail": "catalogue/tool-captures/elevenlabs.jpg", "keywords": [ "Text", "Optional authorised reference audio", "Speech audio", "Default for the established hosted narration route; use voice approval before producing video." ], "thumbnailEvidence": "Official cover image from the ElevenLabs homepage, retrieved 14 September 2026." }, { "id": "tool-voicebox", "title": "Voicebox", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Local voice application", "summary": "Voice studio with cloning, story editing and additional voice input/output features.", "href": "compare/tools/voicebox.html", "thumbnail": "catalogue/tool-captures/voicebox.webp", "keywords": [ "Text", "Reference audio", "Generated speech", "Edited voice stories", "Start here for a local voice-production interface already recorded in the kit." ], "thumbnailEvidence": "Project screenshot from the official source, retrieved 5 September 2026. Not a comparative quality test." }, { "id": "tool-voicestudio", "title": "VoiceStudio", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Multi-engine voice application", "summary": "Multi-engine application for voice creation, transcription and dubbing.", "href": "compare/tools/voicestudio.html", "thumbnail": "catalogue/tool-captures/voicestudio.png", "keywords": [ "Text", "Reference audio", "Video or audio for dubbing", "Speech", "Dubs", "Transcripts", "Choose when you need to compare or route several local engines in one workspace." ], "thumbnailEvidence": "Project screenshot: engine catalogue from the official source, retrieved 5 September 2026. Not a comparative quality test." }, { "id": "tool-voice-pro", "title": "voice-pro", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Audio workflow application", "summary": "Web interface combining recognition, separation and speech-generation engines.", "href": "compare/tools/voice-pro.html", "thumbnail": "catalogue/tool-captures/voice-pro.jpg", "keywords": [ "Audio or video", "Separated vocals", "Subtitles", "Speech", "Use for a bundled audio-cleanup/subtitle workflow rather than as another voice engine." ], "thumbnailEvidence": "Main-page screenshot from the official voice-pro GitHub README, retrieved 14 September 2026." }, { "id": "tool-demucs", "title": "Demucs", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Source-separation model", "summary": "Separates audio sources such as vocals and accompaniment.", "href": "compare/tools/demucs.html", "thumbnail": "catalogue/tool-captures/demucs.jpg", "keywords": [ "Mixed audio", "Separated audio stems", "Use before transcription or mixing when music masks speech." ], "thumbnailEvidence": "Official diagram from the Meta (Facebook Research) Demucs GitHub README, retrieved 14 September 2026." }, { "id": "tool-elevenlabs-dubbing", "title": "ElevenLabs Dubbing", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Hosted dubbing workflow", "summary": "Translates recorded audio/video while retaining speaker characteristics.", "href": "compare/tools/elevenlabs-dubbing.html", "thumbnail": "catalogue/tool-captures/elevenlabs-dubbing.jpg", "keywords": [ "Audio or video", "Optional approved transcript", "Translated dub", "Choose managed dubbing when hosted processing and manual review are acceptable." ], "thumbnailEvidence": "Official share image (\"Dubbing V2 alpha\") from the ElevenLabs Dubbing Studio product page, retrieved 14 September 2026." }, { "id": "tool-elevenlabs-agents", "title": "ElevenLabs Agents", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Hosted agent platform", "summary": "Platform for building conversational voice agents.", "href": "compare/tools/elevenlabs-agents.html", "thumbnail": "catalogue/tool-captures/elevenlabs-agents.jpg", "keywords": [ "Live speech", "Knowledge and tools", "Spoken responses", "Tool actions", "Use for a hosted guide with configured knowledge and tools." ], "thumbnailEvidence": "Official share image from the ElevenLabs Agents app page, retrieved 14 September 2026." }, { "id": "tool-openai-realtime", "title": "OpenAI Realtime", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Realtime model API", "summary": "API for low-latency audio conversations and application tools.", "href": "compare/tools/openai-realtime.html", "thumbnail": "catalogue/tool-captures/openai-realtime.jpg", "keywords": [ "Live audio", "Instructions", "Tool definitions", "Audio/text responses", "Tool calls", "Choose for a custom conversational application where you control the integration." ], "thumbnailEvidence": "GitHub repository social preview card for OpenAI's official openai-realtime-console reference app, retrieved 14 September 2026." }, { "id": "tool-gemini-live", "title": "Gemini Live", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Realtime multimodal API", "summary": "Realtime API for conversational audio and multimodal sessions.", "href": "compare/tools/gemini-live.html", "thumbnail": "catalogue/tool-captures/gemini-live.jpg", "keywords": [ "Streaming audio", "Optional visual context", "Audio/text responses", "Choose when live voice must interact with visual context in a Google-based integration." ], "thumbnailEvidence": "Official share thumbnail from the Gemini app homepage (gemini.google.com), retrieved 14 September 2026." }, { "id": "tool-grok-voice", "title": "Grok Voice Agent API", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Realtime model API", "summary": "Realtime voice interface with configurable tools.", "href": "compare/tools/grok-voice.html", "thumbnail": "catalogue/tool-captures/grok-voice.jpg", "keywords": [ "Live audio", "Tools", "Spoken responses", "Tool calls", "Evaluate when an xAI-based voice integration is specifically useful." ], "thumbnailEvidence": "Official share image from xAI's Grok Voice Agent API documentation page, retrieved 14 September 2026." }, { "id": "tool-pipecat", "title": "Pipecat", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Voice-agent framework", "summary": "Framework connecting realtime transports and voice/AI services.", "href": "compare/tools/pipecat.html", "thumbnail": "catalogue/tool-captures/pipecat.jpg", "keywords": [ "Audio transport", "ASR/LLM/TTS components", "Working voice-agent pipeline", "Use when you need control over model providers and the conversation pipeline." ], "thumbnailEvidence": "Headless capture of the official Pipecat homepage, retrieved 14 September 2026." }, { "id": "tool-hf-speech-to-speech", "title": "Hugging Face speech-to-speech", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Reference application stack", "summary": "Open-source assembly of components for voice agents.", "href": "compare/tools/hf-speech-to-speech.html", "thumbnail": "catalogue/tool-captures/hf-speech-to-speech.jpg", "keywords": [ "Audio", "Models and runtime", "Speech-to-speech pipeline", "Use as a reference or starting implementation for an open-model pipeline." ], "thumbnailEvidence": "Endpoint-swap demo GIF frame from the official Hugging Face speech-to-speech GitHub README, retrieved 14 September 2026." }, { "id": "tool-moshi", "title": "Moshi", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Dialogue model and framework", "summary": "Speech-text model and full-duplex spoken-dialogue research framework.", "href": "compare/tools/moshi.html", "thumbnail": "catalogue/tool-captures/moshi.jpg", "keywords": [ "Live audio", "Spoken dialogue", "Evaluate when research-level control over full-duplex dialogue is the objective." ], "thumbnailEvidence": "Official diagram from the Kyutai Labs Moshi GitHub README, retrieved 14 September 2026." }, { "id": "tool-nemotron-voicechat", "title": "Nemotron VoiceChat", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Dialogue model", "summary": "NVIDIA real-time conversational speech model.", "href": "compare/tools/nemotron-voicechat.html", "thumbnail": "catalogue/tool-captures/nemotron-voicechat.jpg", "keywords": [ "Audio stream", "Conversational audio", "Keep as a candidate for a self-hosted dialogue evaluation." ], "thumbnailEvidence": "Official share image from NVIDIA's Nemotron VoiceChat page on build.nvidia.com, retrieved 14 September 2026." }, { "id": "tool-kokoro", "title": "Kokoro", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech engine or client", "summary": "Compact text-to-speech model with supplied voices.", "href": "compare/tools/kokoro.html", "thumbnail": "catalogue/tool-captures/kokoro.jpg", "keywords": [ "Text", "Reference audio where supported", "Speech audio", "Choose for lightweight speech synthesis using available voices." ], "thumbnailEvidence": "Official Hugging Face social-thumbnail image for the Kokoro-82M model card, retrieved 14 September 2026." }, { "id": "tool-edge-tts", "title": "Edge-TTS", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech engine or client", "summary": "Python client for Microsoft Edge’s online text-to-speech service.", "href": "compare/tools/edge-tts.html", "thumbnail": "catalogue/tool-captures/edge-tts.jpg", "keywords": [ "Text", "Reference audio where supported", "Speech audio", "Use for a simple online speech-generation route when that service dependency is acceptable." ], "thumbnailEvidence": "GitHub repository social preview card for rany2/edge-tts, retrieved 14 September 2026." }, { "id": "tool-f5-tts", "title": "F5-TTS", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech engine or client", "summary": "Flow-matching text-to-speech system for reference-voice generation.", "href": "compare/tools/f5-tts.html", "thumbnail": "catalogue/tool-captures/f5-tts.jpg", "keywords": [ "Text", "Reference audio where supported", "Speech audio", "Evaluate as a self-hosted reference-voice engine inside a supported application." ], "thumbnailEvidence": "GitHub repository social preview card for SWivid/F5-TTS, retrieved 14 September 2026." }, { "id": "tool-cosyvoice", "title": "CosyVoice", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech engine or client", "summary": "Multilingual speech-generation family with inference and deployment tooling.", "href": "compare/tools/cosyvoice.html", "thumbnail": "catalogue/tool-captures/cosyvoice.jpg", "keywords": [ "Text", "Reference audio where supported", "Speech audio", "Choose when you need a controllable multilingual model route in an application you can operate." ], "thumbnailEvidence": "Official image from the QwenAudio CosyVoice GitHub README, retrieved 14 September 2026." }, { "id": "tool-qwen3-tts", "title": "Qwen3-TTS", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech-generation engine", "summary": "Speech generation family with separate voice-design, custom-voice and cloning models.", "href": "compare/tools/qwen3-tts.html", "thumbnail": "catalogue/tool-captures/qwen3-tts.jpg", "keywords": [ "Text", "Reference audio where supported", "Speech audio", "Choose when you need instruction-led voice design or a supported cloning route." ], "thumbnailEvidence": "Official introduction image from the Qwen3-TTS GitHub README (hosted on Alibaba's own CDN), retrieved 14 September 2026." }, { "id": "tool-chatterbox", "title": "Chatterbox", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech-generation engine", "summary": "Open speech-generation model family with voice-reference and expressiveness controls.", "href": "compare/tools/chatterbox.html", "thumbnail": "catalogue/tool-captures/chatterbox.jpg", "keywords": [ "Text", "Reference audio where supported", "Speech audio", "Choose for reference-voice speech generation when its supported version and language fit." ], "thumbnailEvidence": "Official image from the Resemble AI Chatterbox GitHub README, retrieved 14 September 2026." }, { "id": "tool-indextts2", "title": "IndexTTS 2 / 2.5", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech-generation engine", "summary": "Controllable zero-shot speech-generation system.", "href": "compare/tools/indextts2.html", "thumbnail": "catalogue/tool-captures/indextts2.jpg", "keywords": [ "Text", "Reference audio where supported", "Speech audio", "Evaluate when precise control over the generated speech is needed." ], "thumbnailEvidence": "Official video-cover image from the IndexTTS2/2.5 GitHub README, retrieved 14 September 2026." }, { "id": "tool-auk-flash", "title": "AuK-Flash", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Speech-generation engine", "summary": "Tencent's AuK (announced as \"Nano Banana for audio\"), here in its distilled four-step version: speech from a reference voice or a written voice description, plus editing, cleaning and separation of existing speech through written instructions.", "href": "compare/tools/auk-flash.html", "thumbnail": "catalogue/tool-captures/auk-flash.jpg", "keywords": [ "Written instruction with the text to speak", "Reference audio to clone, or source audio to edit", "Speech audio at 24 kHz", "Edited, cleaned or separated recording", "Repairing an existing voice track by instruction (swap one mispronounced word, change emotion, remove breaths or noise) and making extra takes of a speaker's voice from a short reference, in non-commercial work." ], "thumbnailEvidence": "Official Hugging Face social-thumbnail image for Tencent's AuK-Flash model card, retrieved 14 September 2026." }, { "id": "tool-meta-voice-transcribe", "title": "Meta Voice Transcribe", "domain": "Voice", "domains": [ "Voice" ], "kind": "Tool", "layer": "Unresolved product label", "summary": "A name in the existing material whose exact product/API identity needs resolution.", "href": "compare/tools/meta-voice-transcribe.html", "thumbnail": "catalogue/covers/meta-voice-transcribe.svg", "keywords": [ "Unverified", "Unverified", "Keep for follow-up identification." ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-gemini-image", "title": "Gemini image / Nano Banana", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Image engine", "summary": "Generates and edits raster images through Google image-capable models.", "href": "compare/tools/gemini-image.html", "thumbnail": "catalogue/tool-captures/gemini-image.jpg", "keywords": [ "Prompt", "Reference images", "Raster image", "Conversational illustration and reference-based edits" ], "thumbnailEvidence": "Official share image from the Gemini API image-generation documentation page, retrieved 14 September 2026." }, { "id": "tool-gpt-image", "title": "GPT Image 2", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Image engine", "summary": "Generates and edits raster images with text instructions and image inputs.", "href": "compare/tools/gpt-image.html", "thumbnail": "catalogue/covers/gpt-image.svg", "keywords": [ "Prompt", "Reference images", "Raster image", "Detailed prompted illustration and editing" ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-flux2-local", "title": "FLUX.2 dev on RunPod", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Self-hosted image engine", "summary": "Open-weight FLUX image generation operated through the existing GPU workflow.", "href": "compare/tools/flux2-local.html", "thumbnail": "catalogue/tool-captures/flux2-local.jpg", "keywords": [ "Prompt", "Optional references if deployed version supports them", "Raster images", "Repeatable batches through an owned workflow" ], "thumbnailEvidence": "Official Hugging Face social-thumbnail image for the Black Forest Labs FLUX.2-dev (open-weights) model card, retrieved 14 September 2026." }, { "id": "tool-flux2-hosted", "title": "FLUX.2 hosted models", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Hosted image engine", "summary": "BFL endpoints generate images and perform multi-reference edits.", "href": "compare/tools/flux2-hosted.html", "thumbnail": "catalogue/tool-captures/flux2-hosted.jpg", "keywords": [ "Prompt", "Reference images", "Raster images", "Hosted reference-based generation and editing" ], "thumbnailEvidence": "Headless capture of the official Black Forest Labs homepage, retrieved 14 September 2026. Shows BFL's current flagship model lineup, the vendor behind FLUX.2's hosted API." }, { "id": "tool-recraft", "title": "Recraft", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Raster/vector image engine", "summary": "Provides generation, vector generation, vectorization and editing endpoints.", "href": "compare/tools/recraft.html", "thumbnail": "catalogue/tool-captures/recraft.jpg", "keywords": [ "Prompt", "Reference image", "Mask for relevant editing endpoint", "Raster images", "SVG vectors", "Generated vector illustrations and icons" ], "thumbnailEvidence": "Headless capture of the official Recraft homepage, retrieved 14 September 2026." }, { "id": "tool-ideogram", "title": "Ideogram", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Image engine", "summary": "Image generation and editing with typography-oriented features and structured prompts.", "href": "compare/tools/ideogram.html", "thumbnail": "catalogue/tool-captures/ideogram.jpg", "keywords": [ "Text prompt", "Structured prompt", "Image for remix/edit", "Raster images", "Posters, branding studies and compositions with visible text" ], "thumbnailEvidence": "Official share image from the Ideogram homepage, retrieved 14 September 2026." }, { "id": "tool-seedream", "title": "Seedream", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Image engine", "summary": "ByteDance image-generation family exposed through BytePlus and selected providers.", "href": "compare/tools/seedream.html", "thumbnail": "catalogue/tool-captures/seedream.jpg", "keywords": [ "Prompt", "Reference images where model supports them", "Raster images", "Alternative image generation to evaluate on the same brief" ], "thumbnailEvidence": "Official BytePlus share image from the Seedream source page, retrieved 14 September 2026." }, { "id": "tool-midjourney", "title": "Midjourney", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Image engine and editor", "summary": "Generates visual interpretations from text and image references; web editor supports targeted changes.", "href": "compare/tools/midjourney.html", "thumbnail": "catalogue/covers/midjourney.svg", "keywords": [ "Prompt", "Image reference", "Raster images", "Human-led visual exploration" ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-krea", "title": "Krea", "domain": "Images", "domains": [ "Images", "Video" ], "kind": "Tool", "layer": "Image platform and model host", "summary": "Image creation and enhancement platform with a documented developer API.", "href": "compare/tools/krea.html", "thumbnail": "catalogue/tool-captures/krea.jpg", "keywords": [ "Prompt", "Images", "Supported references", "Images", "Generated or edited media", "Exploring image generation and enhancement workflows", "Choose for interactive exploration when its available models and controls fit the task." ], "thumbnailEvidence": "Headless capture of the official Krea homepage, retrieved 14 September 2026." }, { "id": "tool-higgsfield", "title": "Higgsfield", "domain": "Images", "domains": [ "Images", "Video" ], "kind": "Tool", "layer": "Multi-model platform", "summary": "Routes image and video jobs through an agent-accessible platform.", "href": "compare/tools/higgsfield.html", "thumbnail": "catalogue/tool-captures/higgsfield.jpg", "keywords": [ "Prompt", "Reference assets", "Reference images/video/audio as supported", "Images", "Videos depending on model", "Generated or edited media", "Trying several available models through one interface", "Use when trying several hosted models and camera/reference controls in one workflow." ], "thumbnailEvidence": "Capture of the Higgsfield homepage, 14 September 2026." }, { "id": "tool-heliosgen", "title": "HeliosGen", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Workflow wrapper", "summary": "Visual canvas for chaining prompts, reference images, models and automation.", "href": "compare/tools/heliosgen.html", "thumbnail": "catalogue/tool-captures/heliosgen.jpg", "keywords": [ "Prompts", "Reference images", "Workflow", "Provider-generated images and video", "Designing multi-step generation workflows" ], "thumbnailEvidence": "GitHub repository social preview card for SegFault42/HeliosGen, retrieved 14 September 2026." }, { "id": "tool-wangp", "title": "WanGP / Wan2GP", "domain": "Images", "domains": [ "Images", "Video" ], "kind": "Tool", "layer": "Multi-model runner", "summary": "GPU-oriented application supporting image models alongside video models.", "href": "compare/tools/wangp.html", "thumbnail": "catalogue/tool-captures/wangp.jpg", "keywords": [ "Prompt", "Model-specific image inputs", "Prompts", "References", "Selected model weights", "Raster images", "Outputs from the chosen engine", "Operating supported open image models in a shared GPU application", "Use when you want a shared interface over open models on suitable hardware." ], "thumbnailEvidence": "GitHub repository social preview card for deepbeepmeep/Wan2GP, retrieved 14 September 2026." }, { "id": "tool-concept-diagrams", "title": "concept-diagrams", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Studio recipe", "summary": "Chooses a conceptual structure and visual material, then generates and judges illustrations.", "href": "compare/tools/concept-diagrams.html", "thumbnail": "catalogue/covers/concept-diagrams.svg", "keywords": [ "Concept sentence", "Reference style", "Audience", "Concept illustration", "HTML/SVG labels", "An image that explains one concept before words" ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-diagram-design", "title": "diagram-design", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Diagram authoring recipe", "summary": "Writes structured editorial diagrams as self-contained HTML and SVG.", "href": "compare/tools/diagram-design.html", "thumbnail": "catalogue/tool-captures/diagram-design.jpg", "keywords": [ "Facts", "Relationships", "Diagram brief", "HTML", "SVG", "Static architecture, flows, sequences and timelines" ], "thumbnailEvidence": "Architecture-screenshot thumbnail from the official diagram-design GitHub README, retrieved 14 September 2026." }, { "id": "tool-archify", "title": "archify", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Diagram authoring and validation recipe", "summary": "Builds explorable system maps and supports validation and visual exports.", "href": "compare/tools/archify.html", "thumbnail": "catalogue/tool-captures/archify.jpg", "keywords": [ "Repository evidence", "System facts", "Relationships", "HTML", "SVG", "PNG", "WebM", "A system map grounded in repository evidence" ], "thumbnailEvidence": "Official hero image from the archify GitHub README, retrieved 14 September 2026." }, { "id": "tool-mermaid", "title": "Mermaid", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Diagram language and renderer", "summary": "Turns text definitions into diagrams through a JavaScript renderer.", "href": "compare/tools/mermaid.html", "thumbnail": "catalogue/tool-captures/mermaid.jpg", "keywords": [ "Diagram text", "Rendered diagram", "SVG via compatible exporter", "Maintainable flowcharts and sequence diagrams" ], "thumbnailEvidence": "Official header image (sample rendered diagrams) from the Mermaid GitHub README, retrieved 14 September 2026." }, { "id": "tool-fossflow", "title": "FossFLOW", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Isometric diagram editor and renderer", "summary": "Creates isometric infrastructure diagrams with reusable scene objects.", "href": "compare/tools/fossflow.html", "thumbnail": "catalogue/tool-captures/fossflow.jpg", "keywords": [ "Scene JSON", "Nodes", "Connections", "Isometric diagram", "PNG through recorded harness", "Servers, stores and infrastructure shown as an isometric board" ], "thumbnailEvidence": "Demo GIF frame from the official FossFLOW GitHub README, retrieved 14 September 2026." }, { "id": "tool-figma-images", "title": "Figma Design", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Vector and layout editor", "summary": "Edits design objects and exports selected content as image/vector assets.", "href": "compare/tools/figma-images.html", "thumbnail": "catalogue/tool-captures/figma-images.jpg", "keywords": [ "Shapes", "Text", "Images", "Components", "Editable design", "PNG", "JPG", "SVG", "PDF", "Composing editable assets and complete interface layouts" ], "thumbnailEvidence": "Official share image from the Figma homepage, retrieved 14 September 2026." }, { "id": "tool-open-pencil", "title": "OpenPencil", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Programmable design editor", "summary": "Edits design documents through app, CLI and MCP.", "href": "compare/tools/open-pencil.html", "thumbnail": "catalogue/tool-captures/open-pencil.jpg", "keywords": [ "Design file", "Shapes", "Text", "Images", "Editable design", "Exported assets", "AI-driven editable layout and asset production" ], "thumbnailEvidence": "Official application screenshot from the OpenPencil GitHub README, retrieved 14 September 2026." }, { "id": "tool-penpot", "title": "Penpot", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Vector and prototyping editor", "summary": "Design platform with vector editing, layouts and prototyping guidance.", "href": "compare/tools/penpot.html", "thumbnail": "catalogue/tool-captures/penpot.jpg", "keywords": [ "Shapes", "Text", "Images", "Design files", "Editable design", "Exported assets", "Collaborative editable design when a Penpot instance is chosen" ], "thumbnailEvidence": "Official share image from the Penpot help-center user guide, retrieved 14 September 2026." }, { "id": "tool-koboyo", "title": "Koboyo", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Drawing canvas", "summary": "Browser canvas recorded for manual shapes, text and diagram drawing.", "href": "compare/tools/koboyo.html", "thumbnail": "catalogue/tool-captures/koboyo.jpg", "keywords": [ "Shapes", "Text", "Images", "Diagram text as recorded", "Canvas diagram", "Drawing a quick diagram by hand" ], "thumbnailEvidence": "Official share image from the Koboyo homepage, retrieved 14 September 2026." }, { "id": "tool-unsplash", "title": "Unsplash", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Photography source", "summary": "Source of real photographs under its published licence.", "href": "compare/tools/unsplash.html", "thumbnail": "catalogue/covers/unsplash.svg", "keywords": [ "Subject brief", "Selected photo page", "Photograph", "A real-world photograph for a page" ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-animos", "title": "Animos", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Image presentation and video exporter", "summary": "Arranges existing pictures into animated display patterns.", "href": "compare/tools/animos.html", "thumbnail": "catalogue/tool-captures/animos.jpg", "keywords": [ "Existing images", "Animated presentation", "Video export as recorded", "Showing a set of images with a tunnel, globe, spiral or similar pattern" ], "thumbnailEvidence": "Official share image from the Animos homepage, retrieved 14 September 2026." }, { "id": "tool-qwen-image", "title": "Qwen-Image family", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Image engine", "summary": "Image generation and editing family with released model code and weights for specific variants.", "href": "compare/tools/qwen-image.html", "thumbnail": "catalogue/tool-captures/qwen-image.jpg", "keywords": [ "Prompt", "Images for edit models", "Raster images", "Evaluating open image generation and editing" ], "thumbnailEvidence": "Official benchmark image from the Qwen-Image GitHub README (hosted on Alibaba's own CDN), retrieved 14 September 2026." }, { "id": "tool-meta-image", "title": "Meta Muse Image", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Image engine candidate", "summary": "Image capability mentioned in the existing Studio tool register.", "href": "compare/tools/meta-image.html", "thumbnail": "catalogue/covers/meta-image.svg", "keywords": [ "Not verified", "Not verified", "Watchlist pending primary feature verification" ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-logobook", "title": "Logobook", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Visual reference source", "summary": "Gallery of logos, symbols and trademarks.", "href": "compare/tools/logobook.html", "thumbnail": "catalogue/tool-captures/logobook.jpg", "keywords": [ "Research question", "References and notes", "Studying logo structure and visual precedents" ], "thumbnailEvidence": "Headless capture of the official Logobook homepage (its logo directory), retrieved 14 September 2026." }, { "id": "tool-hugeicons", "title": "Hugeicons", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "Icon asset library", "summary": "Library of icons for design and development.", "href": "compare/tools/hugeicons.html", "thumbnail": "catalogue/tool-captures/hugeicons.jpg", "keywords": [ "Icon concept", "Style choice", "Icon assets", "Selecting a consistent icon vocabulary" ], "thumbnailEvidence": "Official share image from the Hugeicons homepage, retrieved 14 September 2026." }, { "id": "tool-veo", "title": "Veo", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Hosted video model", "summary": "Google video-generation model accessed through supported Google APIs and interfaces.", "href": "compare/tools/veo.html", "thumbnail": "catalogue/tool-captures/veo.jpg", "keywords": [ "Prompt", "Reference image where supported", "Generated clip", "Use for a Google-based scripted shot-generation pipeline." ], "thumbnailEvidence": "Crop of the official Google DeepMind Veo product page hero (\"Veo, our leading video generation model\"), retrieved 14 September 2026." }, { "id": "tool-seedance", "title": "Seedance", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Hosted video model", "summary": "ByteDance video-generation family with multimodal reference capabilities in newer releases.", "href": "compare/tools/seedance.html", "thumbnail": "catalogue/tool-captures/seedance.jpg", "keywords": [ "Prompt", "Image/audio/video references supported by the chosen version", "Generated clip", "Evaluate when reference-driven motion and multi-shot storytelling are important." ], "thumbnailEvidence": "Headless capture of the official ByteDance Seed blog post announcing Seedance 2.5, retrieved 14 September 2026." }, { "id": "tool-grok-imagine", "title": "Grok Imagine Video", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Hosted video model", "summary": "xAI video generation, distinct from Grok’s live conversational voice.", "href": "compare/tools/grok-imagine.html", "thumbnail": "catalogue/tool-captures/grok-imagine.jpg", "keywords": [ "Prompt", "Supported image/video references", "Generated video", "Use for the existing Grok-based generation route, then inspect identity, speech and editability." ], "thumbnailEvidence": "Official share image from xAI's Grok Imagine Video documentation page, retrieved 14 September 2026." }, { "id": "tool-minimax-h3", "title": "MiniMax H3", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Video/audio model", "summary": "MiniMax multimodal video model with reference-based generation.", "href": "compare/tools/minimax-h3.html", "thumbnail": "catalogue/tool-captures/minimax-h3.jpg", "keywords": [ "Text", "Supported image/video/audio references", "Generated audiovisual clip", "Evaluate where mixed reference inputs are needed; select API or deployment route explicitly." ], "thumbnailEvidence": "Official share image from MiniMax's own blog post announcing MiniMax H3, retrieved 14 September 2026." }, { "id": "tool-hailuo", "title": "Hailuo", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Video-generation application", "summary": "MiniMax’s user-facing video creation service.", "href": "compare/tools/hailuo.html", "thumbnail": "catalogue/tool-captures/hailuo.jpg", "keywords": [ "Prompt", "Reference material", "Generated clips", "Choose the app when you want an interactive MiniMax workflow." ], "thumbnailEvidence": "Headless capture of the official Hailuo AI (MiniMax) video-creation app homepage, retrieved 14 September 2026." }, { "id": "tool-kling", "title": "Kling", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Video model/platform family", "summary": "Video-generation family named in the existing Studio routes.", "href": "compare/tools/kling.html", "thumbnail": "catalogue/tool-captures/kling.jpg", "keywords": [ "Prompt", "Supported references", "Generated clips", "Keep as an alternative for a like-for-like shot test through a verified provider." ], "thumbnailEvidence": "Official Kling AI demo image (an image-expansion result) served from Kling's own homepage assets, retrieved 14 September 2026. Shows a real generation result, not a benchmark claim." }, { "id": "tool-wan", "title": "Wan video models", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Open video model family", "summary": "Open model family with text, image, speech-to-video and animation variants.", "href": "compare/tools/wan.html", "thumbnail": "catalogue/tool-captures/wan.jpg", "keywords": [ "Prompt", "Image or audio depending on variant", "Generated video", "Start here when you need an open-model experiment on a suitable GPU." ], "thumbnailEvidence": "Computational-efficiency table image from the official Wan2.2 GitHub README, retrieved 14 September 2026." }, { "id": "tool-ltx", "title": "LTX Video / LTX-2", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Open video model family", "summary": "Lightricks video models and associated generation workflows.", "href": "compare/tools/ltx.html", "thumbnail": "catalogue/tool-captures/ltx.jpg", "keywords": [ "Prompt", "Supported references", "Generated clips", "Evaluate alongside Wan with the same shot brief and hardware constraints." ], "thumbnailEvidence": "GitHub repository social preview card for Lightricks/LTX-2, retrieved 14 September 2026." }, { "id": "tool-hunyuan-video", "title": "HunyuanVideo", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Open video model family", "summary": "Tencent open video model family.", "href": "compare/tools/hunyuan-video.html", "thumbnail": "catalogue/tool-captures/hunyuan-video.jpg", "keywords": [ "Prompt", "References supported by the chosen model", "Generated clips", "Use as a candidate where its selected model and deployment meet the project needs." ], "thumbnailEvidence": "GitHub repository social preview card for Tencent-Hunyuan/HunyuanVideo-1.5, retrieved 14 September 2026." }, { "id": "tool-google-ai-studio", "title": "Google AI Studio", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Model-development interface", "summary": "Google interface for exploring and building with its models.", "href": "compare/tools/google-ai-studio.html", "thumbnail": "catalogue/tool-captures/google-ai-studio.jpg", "keywords": [ "Prompts", "Supported media", "Model responses or generated media", "Use to try the supported Google model and prepare an API workflow." ], "thumbnailEvidence": "Official share image from the Google AI Studio product page on ai.google.dev, retrieved 14 September 2026." }, { "id": "tool-heygen", "title": "HeyGen", "domain": "Video", "domains": [ "Video", "Speakers" ], "kind": "Tool", "layer": "Hosted avatar platform", "summary": "Hosted API platform for creating avatar videos and related presenter workflows.", "href": "compare/tools/heygen.html", "thumbnail": "catalogue/tool-captures/heygen.jpg", "keywords": [ "Presenter/avatar", "Text or audio", "Speaking presenter video", "Choose managed avatar production when you want a hosted script-to-presenter route." ], "thumbnailEvidence": "Official share image from the HeyGen homepage, retrieved 14 September 2026." }, { "id": "tool-liveportrait", "title": "LivePortrait", "domain": "Video", "domains": [ "Video", "Speakers" ], "kind": "Tool", "layer": "Portrait-animation model", "summary": "Transfers driving motion to a source portrait with retargeting controls.", "href": "compare/tools/liveportrait.html", "thumbnail": "catalogue/tool-captures/liveportrait.jpg", "keywords": [ "Portrait", "Driving image/video motion", "Animated portrait", "Use when the desired motion already exists in a driving source." ], "thumbnailEvidence": "Showcase GIF frame from the official LivePortrait GitHub README, retrieved 14 September 2026." }, { "id": "tool-musetalk", "title": "MuseTalk", "domain": "Video", "domains": [ "Video", "Speakers" ], "kind": "Tool", "layer": "Lip-sync model", "summary": "Audio-driven lip synchronisation applied to a face image/video route.", "href": "compare/tools/musetalk.html", "thumbnail": "catalogue/tool-captures/musetalk.jpg", "keywords": [ "Face image/video", "Approved audio", "Lip-synchronised video", "Use when you already have the face material and approved speech to synchronise." ], "thumbnailEvidence": "Gradio demo screenshot from the official MuseTalk GitHub README, retrieved 14 September 2026." }, { "id": "tool-latentsync", "title": "LatentSync", "domain": "Video", "domains": [ "Video", "Speakers" ], "kind": "Tool", "layer": "Lip-sync diffusion model", "summary": "Audio-conditioned lip synchronisation for existing video.", "href": "compare/tools/latentsync.html", "thumbnail": "catalogue/tool-captures/latentsync.jpg", "keywords": [ "Face video", "Approved audio", "Lip-synchronised video", "Compare against MuseTalk using the same face clip and approved audio." ], "thumbnailEvidence": "Model architecture diagram from the official ByteDance LatentSync GitHub README, retrieved 14 September 2026." }, { "id": "tool-ffmpeg", "title": "FFmpeg", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Media processing tool", "summary": "Command-line processing for video, audio, filters, subtitles and export.", "href": "compare/tools/ffmpeg.html", "thumbnail": "catalogue/tool-captures/ffmpeg.jpg", "keywords": [ "Existing media", "Timing and filter instructions", "Edited/converted media", "Default for repeatable assembly, transcoding and technical finishing." ], "thumbnailEvidence": "Headless capture of the top of the official FFmpeg homepage, retrieved 14 September 2026." }, { "id": "tool-remotion", "title": "Remotion", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Code-based video renderer", "summary": "Creates videos from React code: text, charts and layouts placed frame by frame, rendered to MP4.", "href": "compare/tools/remotion.html", "thumbnail": "catalogue/tool-captures/remotion.png", "keywords": [ "Code", "Images", "Audio", "Data", "Rendered video", "Choose for repeatable explainers with precise text, charts and branded composition." ], "thumbnailEvidence": "Frame from a three-second test render made with Remotion's blank starter on 14 September 2026. Shows the kind of output, not a quality comparison." }, { "id": "tool-final-cut-pro", "title": "Final Cut Pro", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Timeline editor", "summary": "Apple video-editing application for assembling and finishing a film.", "href": "compare/tools/final-cut-pro.html", "thumbnail": "catalogue/tool-captures/final-cut-pro.jpg", "keywords": [ "Video", "Audio", "Titles", "Edited film", "Use when a person needs precise editorial control over timing, picture and sound." ], "thumbnailEvidence": "Official share image from Apple's Final Cut Pro product page, retrieved 14 September 2026." }, { "id": "tool-openshot", "title": "OpenShot", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Timeline editor", "summary": "Open-source video editor with timeline editing and effects.", "href": "compare/tools/openshot.html", "thumbnail": "catalogue/tool-captures/openshot.jpg", "keywords": [ "Video", "Audio", "Images", "Edited video", "Choose for accessible manual editing without adopting a commercial editor." ], "thumbnailEvidence": "Official share image from the OpenShot features page, retrieved 14 September 2026." }, { "id": "tool-capcut", "title": "CapCut", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Timeline/content editor", "summary": "Desktop video editor with editing, effects, search and AI-assisted content features.", "href": "compare/tools/capcut.html", "thumbnail": "catalogue/tool-captures/capcut.jpg", "keywords": [ "Video", "Audio", "Text", "Edited video", "Choose for quick manual editing with effects and social-format delivery." ], "thumbnailEvidence": "Official share image from the CapCut desktop video editor page, retrieved 14 September 2026." }, { "id": "tool-real-esrgan", "title": "Real-ESRGAN", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Restoration model", "summary": "Image/video restoration and upscaling project.", "href": "compare/tools/real-esrgan.html", "thumbnail": "catalogue/tool-captures/real-esrgan.jpg", "keywords": [ "Existing frames/images", "Restored/upscaled frames", "Use when existing footage needs restoration, then inspect whether details remain faithful." ], "thumbnailEvidence": "Teaser comparison image from the official Real-ESRGAN GitHub README, retrieved 14 September 2026." }, { "id": "tool-moneyprinterturbo", "title": "MoneyPrinterTurbo", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Automated video workflow", "summary": "Builds short videos through a pipeline of scripts, assets, narration and assembly.", "href": "compare/tools/moneyprinterturbo.html", "thumbnail": "catalogue/tool-captures/moneyprinterturbo.jpg", "keywords": [ "Topic or script", "Providers and source assets", "Assembled short video", "Use for prototyping a repeatable short-video pipeline with reviewed inputs." ], "thumbnailEvidence": "WebUI screenshot from the official MoneyPrinterTurbo GitHub README, retrieved 14 September 2026." }, { "id": "tool-runpod", "title": "RunPod", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "GPU hosting service", "summary": "Infrastructure for running GPU workloads.", "href": "compare/tools/runpod.html", "thumbnail": "catalogue/tool-captures/runpod.jpg", "keywords": [ "Container/runtime", "Model", "Inputs", "Compute execution and model outputs", "Use when an open model needs hardware beyond the local computer." ], "thumbnailEvidence": "Official share image from the RunPod homepage, retrieved 14 September 2026." }, { "id": "tool-yt-dlp", "title": "yt-dlp", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Media acquisition tool", "summary": "Command-line downloader for supported media sources.", "href": "compare/tools/yt-dlp.html", "thumbnail": "catalogue/tool-captures/yt-dlp.jpg", "keywords": [ "Supported media URL", "Local media file", "Use to obtain permitted source media for analysis or editing." ], "thumbnailEvidence": "GitHub repository social preview card for yt-dlp/yt-dlp, retrieved 14 September 2026." }, { "id": "tool-magic-hour", "title": "Magic Hour", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Hosted media platform", "summary": "Hosted platform grouping text/image/video generation and editing workflows.", "href": "compare/tools/magic-hour.html", "thumbnail": "catalogue/tool-captures/magic-hour.jpg", "keywords": [ "Task-specific source material", "Depends on the identified workflow", "Choose a focused hosted task such as photo-to-video or restyling when it fits your source material." ], "thumbnailEvidence": "Official share image from the Magic Hour homepage, retrieved 14 September 2026." }, { "id": "tool-palmier", "title": "Palmier Pro", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Video workflow application", "summary": "Mac timeline editor with generation and agent editing through MCP.", "href": "compare/tools/palmier.html", "thumbnail": "catalogue/tool-captures/palmier.png", "keywords": [ "Task-specific source material", "Depends on the identified workflow", "Evaluate for human-and-agent timeline editing if the machine meets its requirements." ], "thumbnailEvidence": "Project screenshot from the official source, retrieved 5 September 2026. Not a comparative quality test." }, { "id": "tool-wondershare", "title": "Wondershare (product unresolved)", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Vendor label", "summary": "Vendor name in the existing material; the exact editor/product is not specified.", "href": "compare/tools/wondershare.html", "thumbnail": "catalogue/covers/wondershare.svg", "keywords": [ "Task-specific source material", "Depends on the identified workflow", "Retain as a reference; resolve current capability and access before selection." ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-video-toolkit", "title": "Studio video toolkit", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Internal orchestration recipe", "summary": "The local collection of scripts and skills that connects production steps.", "href": "compare/tools/video-toolkit.html", "thumbnail": "catalogue/covers/video-toolkit.svg", "keywords": [ "Task-specific source material", "Depends on the identified workflow", "Retain as a reference; resolve current capability and access before selection." ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-stillslab", "title": "StillsLab", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Visual reference library", "summary": "Film and television frame library for selecting composition, lighting and visual references.", "href": "compare/tools/stillslab.html", "thumbnail": "catalogue/tool-captures/stillslab.jpg", "keywords": [ "Task-specific source material", "Depends on the identified workflow", "Use to choose and describe a visual direction before generating a shot." ], "thumbnailEvidence": "Official share image from the StillsLab homepage, retrieved 14 September 2026." }, { "id": "tool-motionsites", "title": "MotionSites", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Reference/prompt library", "summary": "Website design and motion-prompt reference library.", "href": "compare/tools/motionsites.html", "thumbnail": "catalogue/tool-captures/motionsites.jpg", "keywords": [ "Task-specific source material", "Depends on the identified workflow", "Use for motion-led website references and prompts; keep film production in its own workflow." ], "thumbnailEvidence": "Official share image from the MotionSites homepage, retrieved 14 September 2026." }, { "id": "tool-hand-drawn-explainer-nikola", "title": "Hand-drawn explainer (Nikola)", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Agent skill for explainer videos", "summary": "A Codex skill that turns a topic, a script or a subtitle file into an explainer video drawn stroke by stroke, or animated as cards and diagrams.", "href": "compare/tools/hand-drawn-explainer-nikola.html", "thumbnail": "catalogue/covers/hand-drawn-explainer-nikola.svg", "keywords": [ "Topic", "Script", "SRT subtitles", "Source picture", "MP4 video", "Subtitles", "Timeline", "Editable project", "Choose when an idea should be drawn in front of the viewer, one part at a time, with subtitles and an editable project." ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-video-shotcraft", "title": "Video Shotcraft", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Agent skill for product films", "summary": "A Claude Code and Codex skill that turns a product's real screens into a short promo film on Remotion, using 157 shot recipes with tuned code, sound effects and optional music.", "href": "compare/tools/video-shotcraft.html", "thumbnail": "catalogue/tool-captures/video-shotcraft.jpg", "keywords": [ "Product URL or local server", "Screenshots", "Copy and brand styles", "Music (optional)", "MP4 with and without music", "Remotion project", "Browser-editable timeline", "JianYing draft (optional)", "Choose for a product or service film built from its own screens, when exact text and an editable result matter more than a speaker." ], "thumbnailEvidence": "Capture of the project's online gallery, 14 September 2026." }, { "id": "tool-infinitetalk", "title": "InfiniteTalk", "domain": "Video", "domains": [ "Video", "Speakers" ], "kind": "Tool", "layer": "Open audio-driven avatar model", "summary": "Turns a portrait or a clip and an approved audio track into a speaking video with head and body motion. Open weights from MeiGen-AI.", "href": "compare/tools/infinitetalk.html", "thumbnail": "catalogue/tool-captures/infinitetalk.jpg", "keywords": [ "Portrait image or clip", "Approved audio", "Optional prompt", "Speaking video, 480p or 720p", "The cheapest measured route for a portrait speaking our own audio: identity held, about $2.30 a published minute at 480p on WaveSpeed." ], "thumbnailEvidence": "Frame from an official demo video in the InfiniteTalk GitHub README (a model output, its watermark kept), retrieved 15 September 2026." }, { "id": "tool-longcat-video-avatar", "title": "LongCat-Video-Avatar 1.5", "domain": "Video", "domains": [ "Video", "Speakers" ], "kind": "Tool", "layer": "Open audio-driven avatar model", "summary": "Turns one portrait and an approved audio track into a speaking video, face and body born together. Open weights from Meituan.", "href": "compare/tools/longcat-video-avatar.html", "thumbnail": "catalogue/tool-captures/longcat-video-avatar.jpg", "keywords": [ "Portrait image", "Approved audio", "Optional prompt", "Speaking video, 480p or 720p, up to 2 minutes a job on WaveSpeed", "A portrait speaking our own approved audio, when InfiniteTalk misses lip shapes." ], "thumbnailEvidence": "Frame from the official demo clip assets/2.mp4 on the LongCat-Video-Avatar-1.5 model card (a model output), retrieved 15 September 2026." }, { "id": "tool-video-use", "title": "video-use", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Claude Code editing skill", "summary": "A Claude Code skill that edits talking footage from its transcript: picks takes, cuts filler words and dead air, grades, subtitles and renders with ffmpeg.", "href": "compare/tools/video-use.html", "thumbnail": "catalogue/tool-captures/video-use.jpg", "keywords": [ "Folder of raw video takes", "A plain-English brief", "edit/final.mp4 with optional subtitles, colour grade and overlays", "A cut list (edl.json) and a project.md memory", "Raw takes of people speaking (a recorded speaker, a demo, an interview) cut into one clean film by asking Claude." ], "thumbnailEvidence": "Official banner from the video-use GitHub repository: a Claude Code session running the skill's task list, retrieved 15 September 2026." }, { "id": "tool-flapkit", "title": "Flapkit", "domain": "UI", "domains": [ "UI" ], "kind": "Tool", "layer": "React component library", "summary": "Split-flap departure board for React: letters and digits flip over to their new value, with sound.", "href": "compare/tools/flapkit.html", "thumbnail": "catalogue/tool-captures/flapkit.jpg", "keywords": [ "Short strings, one character per cell", "A character deck, built in or your own", "A motion adapter: cascade or riffle, CSS or Canvas", "An animated split-flap board inside the page", "Mechanical flip sound, optional", "Choose inside a React product when a short value must be seen changing, the way an airport board changes." ], "thumbnailEvidence": "Official documentation demo at cuvii.dev/flapkit, captured 17 September 2026." }, { "id": "tool-cube-motion", "title": "Cube", "domain": "UI", "domains": [ "UI" ], "kind": "Tool", "layer": "JavaScript animation library", "summary": "Four ready motions for page elements: rise (entrance), leave (exit), morph (a label or icon changing state) and reveal (rise when scrolled into view). Durations, curves and staggers are fixed by the library.", "href": "compare/tools/cube-motion.html", "thumbnail": "catalogue/tool-captures/cube-motion.jpg", "keywords": [ "The elements to move, by selector or node", "Optional: animate the children one after another", "Entrance, exit, state-change and on-scroll motion inside the page", "Choose when a page needs calm entrances, exits and state changes without tuning durations and curves by hand, including on a house single-file page." ], "thumbnailEvidence": "Official site at cube-motion.dev, captured 21 September 2026." }, { "id": "tool-map3d", "title": "map3d", "domain": "Images", "domains": [ "Images" ], "kind": "Tool", "layer": "3D map generator (web app, open source)", "summary": "Turns any area of the world into a 3D city map: buildings extruded to their recorded height and roads, from OpenStreetMap, exportable as a GLB 3D model.", "href": "compare/tools/map3d.html", "thumbnail": "catalogue/tool-captures/map3d.jpg", "keywords": [ "An area on the world map", "OpenStreetMap building and road data, fetched by the app", "A 3D city map in the browser", "A GLB 3D model file for Blender, three.js, a page or a film", "Choose for a 3D view of a real city, district, port or corridor that a page or a film can show, turn or fly over." ], "thumbnailEvidence": "Project screenshot from the official README, retrieved 21 September 2026." }, { "id": "tool-hyperframes", "title": "HyperFrames", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Code-based video renderer", "summary": "HeyGen's open-source framework that turns an HTML page with timed animations into an MP4. Claude writes each scene as a web page and renders it; a change edits one line and re-renders.", "href": "compare/tools/hyperframes.html", "thumbnail": "catalogue/covers/hyperframes.svg", "keywords": [ "HTML, CSS and JavaScript scenes", "Images", "Video", "Audio", "Rendered MP4 video", "Motion-graphics explainers and product films when the scenes are closer to web pages than to video, especially reusing a site's own HTML and CSS." ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-21st-dev", "title": "21st", "domain": "UI", "domains": [ "UI" ], "kind": "Tool", "layer": "Interface component library", "summary": "A library of over 10,000 React and Tailwind interface components (buttons, cards, menus, sections) made by design engineers, searchable from Claude through its own connector.", "href": "compare/tools/21st-dev.html", "thumbnail": "catalogue/covers/21st-dev.svg", "keywords": [ "A description of the component you need", "React and Tailwind component code", "Showing a believable product screen inside a motion-graphics film, or a React page that needs polished parts instead of improvised ones." ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." }, { "id": "tool-whatships", "title": "What Ships", "domain": "Video", "domains": [ "Video" ], "kind": "Tool", "layer": "Reference video catalogue", "summary": "A free, curated catalogue of about 2,200 startup launch and demo videos posted on X, sorted by category, with duration and poster frames.", "href": "compare/tools/whatships.html", "thumbnail": "catalogue/covers/whatships.svg", "keywords": [ "A category or product type", "Links to reference videos", "Picking one or two motion-graphics films whose pace, type and transitions Claude should copy before it builds a video." ], "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result." } ], "tools": [ { "id": "freeflow", "name": "FreeFlow", "domain": "Voice", "kind": "Dictation application", "compareTask": "Dictate into an app", "summary": "Mac dictation with transcription, cleanup and spoken text editing.", "bestFor": "Default for the existing dictation workflow; choose the backend deliberately.", "notFor": "It is not a speech-generation model.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Mac app with Groq or OpenAI-compatible providers", "inputs": [ "Microphone speech" ], "outputs": [ "Text in the active app" ], "limitations": [ "Cloud use depends on backend; local providers are also supported." ], "sourceUrls": [ { "title": "FreeFlow official documentation", "url": "https://github.com/zachlatta/freeflow/blob/main/README.md" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Dictate into an app", "stars": 5, "rationale": "Existing recorded daily workflow plus configurable providers. This is a workflow recommendation, not a claim of best recognition accuracy.", "recommendation": "Default for the existing dictation workflow; choose the backend deliberately.", "domain": "Voice", "compareTask": "Dictate into an app" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/freeflow.png", "thumbnailEvidence": "Official FreeFlow application icon. Open the project source for its demonstration.", "comparisonNotes": "It is not a speech-generation model. Cloud use depends on backend; local providers are also supported.", "costModel": "Free application; provider or local-compute costs depend on configuration.", "dataLocation": "Configured provider: local/self-hosted or cloud.", "languageSupport": "Language coverage follows the selected recognition backend.", "domains": [ "Voice" ], "thumbnailSource": "https://raw.githubusercontent.com/zachlatta/freeflow/main/Resources/AppIcon-Source.png" }, { "id": "fluidvoice", "name": "FluidVoice", "domain": "Voice", "kind": "Dictation application", "compareTask": "Dictate into an app", "summary": "Local-first dictation with selectable speech engines and optional enhancement.", "bestFor": "Choose when local processing is the priority; compare your own vocabulary against FreeFlow.", "notFor": "Not automatically cloud-free in every configuration.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Mac app; local models or optional cloud enhancement", "inputs": [ "Microphone speech", "Recorded audio" ], "outputs": [ "Dictated text", "File transcripts" ], "limitations": [ "Confirm language/model support and speaker-label output in the installed version." ], "sourceUrls": [ { "title": "FluidVoice official documentation", "url": "https://github.com/altic-dev/FluidVoice" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Dictate into an app", "stars": 4, "rationale": "A strong local-first alternative. Engine and enhancement settings, rather than the app name, determine privacy and results.", "recommendation": "Choose when local processing is the priority; compare your own vocabulary against FreeFlow.", "domain": "Voice", "compareTask": "Dictate into an app" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/fluidvoice.png", "thumbnailEvidence": "Project screenshot: file transcription from the official source, retrieved 5 September 2026. Not a comparative quality test.", "comparisonNotes": "Not automatically cloud-free in every configuration. Confirm language/model support and speaker-label output in the installed version.", "costModel": "Application/local-compute costs; optional providers have their own charges.", "dataLocation": "Local-first; optional cloud enhancement must be checked.", "languageSupport": "Depends on the installed ASR engine and version.", "domains": [ "Voice" ], "thumbnailSource": "https://raw.githubusercontent.com/altic-dev/FluidVoice/main/.github/screenshots/file-transcription-speaker-labelled-result.png" }, { "id": "wispr-flow", "name": "Wispr Flow", "domain": "Voice", "kind": "Dictation application", "compareTask": "Dictate into an app", "summary": "Commercial dictation product already considered in the personal workflow.", "bestFor": "Reconsider only if a new trial fixes the recognition problems previously reported.", "notFor": "Not the default for Frank based on his recorded feedback.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Desktop/mobile product; plan and platform vary", "inputs": [ "Speech" ], "outputs": [ "Dictated text" ], "limitations": [ "Current version has not been retested; this rating is based on recorded user preference." ], "sourceUrls": [ { "title": "Wispr Flow official documentation", "url": "https://wisprflow.ai/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Recorded user verdict; current quality untested", "ratings": [ { "task": "Dictate into an app", "stars": 1, "rationale": "Personal suitability only: Frank previously rejected its recognition. No claim about other users or the current release.", "recommendation": "Reconsider only if a new trial fixes the recognition problems previously reported.", "domain": "Voice", "compareTask": "Dictate into an app" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/wispr-flow.jpg", "thumbnailEvidence": "Official product preview from the official source, retrieved 5 September 2026. Not a comparative quality test.", "comparisonNotes": "Not the default for Frank based on his recorded feedback. Current version has not been retested; this rating is based on recorded user preference.", "costModel": "Service plan; check the current tier.", "dataLocation": "Check the service and privacy settings before sensitive dictation.", "languageSupport": "Check current product coverage and recognition on your vocabulary.", "domains": [ "Voice" ], "thumbnailSource": "https://cdn.prod.website-files.com/682f84b3838c89f8ff7667db/6a724a4fff322698a57a7d97_dictation-og%20(1).jpg" }, { "id": "whisper", "name": "Whisper", "domain": "Voice", "kind": "Speech-recognition model", "compareTask": "Transcribe a recording", "summary": "Open speech-recognition model for transcription, translation and language identification.", "bestFor": "Use as a general multilingual transcription engine or baseline.", "notFor": "Speaker identification requires a separate diarisation layer.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Local runtime or a hosted service running Whisper", "inputs": [ "Audio file" ], "outputs": [ "Transcript", "Timed segments" ], "limitations": [ "Check names and technical terms; distinguish the open model from hosted transcription products." ], "sourceUrls": [ { "title": "Whisper official documentation", "url": "https://github.com/openai/whisper" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Transcribe a recording", "stars": 4, "rationale": "A clear, reusable transcription baseline. Language and hardware affect results; no universal accuracy ranking is implied.", "recommendation": "Use as a general multilingual transcription engine or baseline.", "domain": "Voice", "compareTask": "Transcribe a recording" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/whisper.jpg", "thumbnailEvidence": "Architecture diagram from the official OpenAI Whisper GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Speaker identification requires a separate diarisation layer. Check names and technical terms; distinguish the open model from hosted transcription products.", "costModel": "Local compute, or hosted-service charges.", "dataLocation": "Local when run locally; cloud when used through a hosted endpoint.", "languageSupport": "Multilingual; accuracy varies by language and model.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/openai/whisper" }, { "id": "groq-whisper", "name": "Groq speech-to-text", "domain": "Voice", "kind": "Hosted inference service", "compareTask": "Transcribe a recording", "summary": "Hosted transcription endpoint used by existing Studio ingestion workflows.", "bestFor": "Use when cloud processing is acceptable and an existing API workflow is useful.", "notFor": "Not a separate Whisper model family or an offline route.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Hosted API", "inputs": [ "Audio file" ], "outputs": [ "Transcript" ], "limitations": [ "Audio leaves the local device; service quotas and supported models must be checked." ], "sourceUrls": [ { "title": "Groq speech-to-text official documentation", "url": "https://console.groq.com/docs/speech-to-text" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Transcribe a recording", "stars": 4, "rationale": "Practical fit for the recorded automated ingestion route; no current speed or cost claim.", "recommendation": "Use when cloud processing is acceptable and an existing API workflow is useful.", "domain": "Voice", "compareTask": "Transcribe a recording" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/groq-whisper.jpg", "thumbnailEvidence": "Headless capture of the official Groq homepage, retrieved 14 September 2026. Groq speech-to-text is one of Groq's hosted API models.", "comparisonNotes": "Not a separate Whisper model family or an offline route. Audio leaves the local device; service quotas and supported models must be checked.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Voice" ], "thumbnailSource": "https://groq.com/" }, { "id": "parakeet", "name": "Parakeet TDT", "domain": "Voice", "kind": "Speech-recognition model", "compareTask": "Transcribe a recording", "summary": "NVIDIA speech-recognition model used through compatible apps or runtimes.", "bestFor": "Use through FluidVoice or a supported runtime when the selected model covers the language.", "notFor": "Not a complete dictation app; do not assume all Parakeet versions support the same languages.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Model runtime or host application", "inputs": [ "Audio" ], "outputs": [ "Transcript" ], "limitations": [ "Select the version explicitly; timestamps and formatting depend on the integration." ], "sourceUrls": [ { "title": "Parakeet TDT official documentation", "url": "https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Transcribe a recording", "stars": 4, "rationale": "A focused ASR alternative; compare the same recording and language rather than vendor headline benchmarks.", "recommendation": "Use through FluidVoice or a supported runtime when the selected model covers the language.", "domain": "Voice", "compareTask": "Transcribe a recording" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/parakeet.jpg", "thumbnailEvidence": "Official Hugging Face social-thumbnail image for the NVIDIA Parakeet TDT model card, retrieved 14 September 2026.", "comparisonNotes": "Not a complete dictation app; do not assume all Parakeet versions support the same languages. Select the version explicitly; timestamps and formatting depend on the integration.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3" }, { "id": "vibeasr-cpp", "name": "VibeASR.cpp / BitNet", "domain": "Voice", "kind": "CPU inference runtime", "compareTask": "Transcribe a recording", "summary": "CPU runtime for the compressed VibeVoice-ASR-BitNet model.", "bestFor": "Use for the recorded local transcription route when a plain transcript is sufficient.", "notFor": "Do not inherit speaker labels and timestamps from another VibeVoice release.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Local CPU runtime", "inputs": [ "Recorded audio" ], "outputs": [ "Transcript" ], "limitations": [ "The private measured record remains authoritative for actual output fields and performance." ], "sourceUrls": [ { "title": "VibeASR.cpp / BitNet official documentation", "url": "https://github.com/microsoft/VibeASR.cpp" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official runtime identity plus existing local-use record", "ratings": [ { "task": "Transcribe a recording", "stars": 4, "rationale": "The existing local record demonstrates a useful route. Output expectations must follow that runtime, not the broader family.", "recommendation": "Use for the recorded local transcription route when a plain transcript is sufficient.", "domain": "Voice", "compareTask": "Transcribe a recording" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/vibeasr-cpp.jpg", "thumbnailEvidence": "Report-overview image from the official Microsoft VibeASR.cpp GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Do not inherit speaker labels and timestamps from another VibeVoice release. The private measured record remains authoritative for actual output fields and performance.", "costModel": "Local CPU compute and storage.", "dataLocation": "Local execution.", "languageSupport": "Multilingual; measured local output remains in the private kit.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/microsoft/VibeASR.cpp" }, { "id": "vibevoice-asr", "name": "VibeVoice-ASR", "domain": "Voice", "kind": "Speech-recognition model", "compareTask": "Speaker-labelled transcription", "summary": "Model with documented structured transcription, speaker and timing output.", "bestFor": "Evaluate when who-spoke-when output matters.", "notFor": "Not the same deployment as the CPU BitNet runtime already used locally.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Model/runtime deployment", "inputs": [ "Long audio recording" ], "outputs": [ "Structured transcript", "Speaker labels", "Timestamps" ], "limitations": [ "Do not advertise structured output as proven locally. Streaming ASR is a separate release." ], "sourceUrls": [ { "title": "VibeVoice-ASR official documentation", "url": "https://github.com/microsoft/VibeVoice" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Speaker-labelled transcription", "stars": 3, "rationale": "Good capability fit on paper, but the required output needs verification in the chosen runtime before becoming the default.", "recommendation": "Evaluate when who-spoke-when output matters.", "domain": "Voice", "compareTask": "Speaker-labelled transcription" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/vibevoice-asr.jpg", "thumbnailEvidence": "Diarization-error-rate (DER) benchmark chart from the official Microsoft VibeVoice GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Not the same deployment as the CPU BitNet runtime already used locally. Do not advertise structured output as proven locally. Streaming ASR is a separate release.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/microsoft/VibeVoice" }, { "id": "vibevoice-streaming", "name": "VibeVoice-ASR-Streaming", "domain": "Voice", "kind": "Streaming recognition model", "compareTask": "Speaker-labelled transcription", "summary": "Recently released streaming branch of the VibeVoice family.", "bestFor": "Evaluate for live transcription where streaming is necessary.", "notFor": "Not a reason to replace a working file-transcription route without testing.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Streaming ASR integration", "inputs": [ "Incoming audio stream" ], "outputs": [ "Incremental transcript" ], "limitations": [ "Language coverage and hotword behaviour differ by release; untested here." ], "sourceUrls": [ { "title": "VibeVoice-ASR-Streaming official documentation", "url": "https://github.com/microsoft/VibeVoice" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Speaker-labelled transcription", "stars": null, "rationale": "New release identified; no local suitability evidence yet.", "recommendation": "Evaluate for live transcription where streaming is necessary.", "domain": "Voice", "compareTask": "Speaker-labelled transcription" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/vibevoice-streaming.jpg", "thumbnailEvidence": "tcpWER benchmark chart from the official Microsoft VibeVoice GitHub README, retrieved 14 September 2026 (a different chart than the one used for VibeVoice-ASR).", "comparisonNotes": "Not a reason to replace a working file-transcription route without testing. Language coverage and hotword behaviour differ by release; untested here.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/microsoft/VibeVoice" }, { "id": "elevenlabs", "name": "ElevenLabs", "domain": "Voice", "kind": "Hosted voice platform", "compareTask": "Generate narration and cloned speech", "summary": "Hosted speech generation, voice design, cloning and other audio capabilities.", "bestFor": "Default for the established hosted narration route; use voice approval before producing video.", "notFor": "Not an offline model; audio narration does not create a talking face.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Web application and API", "inputs": [ "Text", "Optional authorised reference audio" ], "outputs": [ "Speech audio" ], "limitations": [ "Choose the appropriate product: narration, dubbing, transcription and agents are different tasks. Review the output by ear." ], "sourceUrls": [ { "title": "ElevenLabs official documentation", "url": "https://elevenlabs.io/docs/overview/intro/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Generate narration and cloned speech", "stars": 5, "rationale": "Fits the existing Studio narration workflow and exposes a documented API. Five stars means this workflow default, not the best voice in a blind test.", "recommendation": "Default for the established hosted narration route; use voice approval before producing video.", "domain": "Voice", "compareTask": "Generate narration and cloned speech" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/elevenlabs.jpg", "thumbnailEvidence": "Official cover image from the ElevenLabs homepage, retrieved 14 September 2026.", "comparisonNotes": "Not an offline model; audio narration does not create a talking face. Choose the appropriate product: narration, dubbing, transcription and agents are different tasks. Review the output by ear.", "costModel": "Hosted plan and usage allowances.", "dataLocation": "Hosted voice processing.", "languageSupport": "Select a supported model and test pronunciation in the target language.", "domains": [ "Voice" ], "thumbnailSource": "https://elevenlabs.io/" }, { "id": "voicebox", "name": "Voicebox", "domain": "Voice", "kind": "Local voice application", "compareTask": "Generate narration and cloned speech", "summary": "Voice studio with cloning, story editing and additional voice input/output features.", "bestFor": "Start here for a local voice-production interface already recorded in the kit.", "notFor": "Do not confuse it with Meta’s unrelated Voicebox research model.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Desktop application and integrations", "inputs": [ "Text", "Reference audio" ], "outputs": [ "Generated speech", "Edited voice stories" ], "limitations": [ "The upstream app has expanded beyond TTS. Check configured backends before promising fully local operation." ], "sourceUrls": [ { "title": "Voicebox official documentation", "url": "https://github.com/jamiepine/voicebox" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Generate narration and cloned speech", "stars": 4, "rationale": "A local application is easier to adopt than assembling engines directly; exact features depend on the installed release.", "recommendation": "Start here for a local voice-production interface already recorded in the kit.", "domain": "Voice", "compareTask": "Generate narration and cloned speech" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/voicebox.webp", "thumbnailEvidence": "Project screenshot from the official source, retrieved 5 September 2026. Not a comparative quality test.", "comparisonNotes": "Do not confuse it with Meta’s unrelated Voicebox research model. The upstream app has expanded beyond TTS. Check configured backends before promising fully local operation.", "costModel": "Application plus selected model/compute costs.", "dataLocation": "Check local model selection and any connected services.", "languageSupport": "Selected engine and voice determine coverage.", "domains": [ "Voice" ], "thumbnailSource": "https://raw.githubusercontent.com/jamiepine/voicebox/main/landing/public/assets/app-screenshot-1.webp" }, { "id": "voicestudio", "name": "VoiceStudio", "domain": "Voice", "kind": "Multi-engine voice application", "compareTask": "Generate narration and cloned speech", "summary": "Multi-engine application for voice creation, transcription and dubbing.", "bestFor": "Choose when you need to compare or route several local engines in one workspace.", "notFor": "Its engine catalogue is not a guarantee that every engine fits the hardware or licence needs.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application with separately selected engines", "inputs": [ "Text", "Reference audio", "Video or audio for dubbing" ], "outputs": [ "Speech", "Dubs", "Transcripts" ], "limitations": [ "Capabilities and licensing depend on the selected engine and weights; language catalogue size is not equal tested quality." ], "sourceUrls": [ { "title": "VoiceStudio official documentation", "url": "https://github.com/debpalash/VoiceStudio" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Generate narration and cloned speech", "stars": 4, "rationale": "Broader workflow coverage than a single TTS engine, with more configuration and model choices.", "recommendation": "Choose when you need to compare or route several local engines in one workspace.", "domain": "Voice", "compareTask": "Generate narration and cloned speech" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/voicestudio.png", "thumbnailEvidence": "Project screenshot: engine catalogue from the official source, retrieved 5 September 2026. Not a comparative quality test.", "comparisonNotes": "Its engine catalogue is not a guarantee that every engine fits the hardware or licence needs. Capabilities and licensing depend on the selected engine and weights; language catalogue size is not equal tested quality.", "costModel": "Local compute and storage; engine/weight terms vary.", "dataLocation": "Local engines, unless connected to a remote backend.", "languageSupport": "A language catalogue is not a quality guarantee for each engine.", "domains": [ "Voice" ], "thumbnailSource": "https://raw.githubusercontent.com/debpalash/VoiceStudio/main/docs/media/0.5.0/catalogue.png" }, { "id": "voice-pro", "name": "voice-pro", "domain": "Voice", "kind": "Audio workflow application", "compareTask": "Isolate voice from mixed audio", "summary": "Web interface combining recognition, separation and speech-generation engines.", "bestFor": "Use for a bundled audio-cleanup/subtitle workflow rather than as another voice engine.", "notFor": "Not one model, and not all bundled routes are offline.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Gradio interface around several tools", "inputs": [ "Audio or video" ], "outputs": [ "Separated vocals", "Subtitles", "Speech" ], "limitations": [ "Edge-TTS is an online client; installing the bundle does not make every feature local." ], "sourceUrls": [ { "title": "voice-pro official documentation", "url": "https://github.com/abus-aikorea/voice-pro" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Isolate voice from mixed audio", "stars": 4, "rationale": "Choose the integrated interface when you want vocal separation plus surrounding audio tools. Demucs is an underlying separation engine, not a separate competing app.", "recommendation": "Use the interface for an assisted isolation workflow; use Demucs directly for a scripted separation step.", "domain": "Voice", "compareTask": "Isolate voice from mixed audio" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/voice-pro.jpg", "thumbnailEvidence": "Main-page screenshot from the official voice-pro GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Not one model, and not all bundled routes are offline. Edge-TTS is an online client; installing the bundle does not make every feature local.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/abus-aikorea/voice-pro" }, { "id": "demucs", "name": "Demucs", "domain": "Voice", "kind": "Source-separation model", "compareTask": "Isolate voice from mixed audio", "summary": "Separates audio sources such as vocals and accompaniment.", "bestFor": "Use before transcription or mixing when music masks speech.", "notFor": "It does not transcribe or generate narration.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Local model or an application such as voice-pro", "inputs": [ "Mixed audio" ], "outputs": [ "Separated audio stems" ], "limitations": [ "Separation can introduce artefacts; it cannot recover every obscured word." ], "sourceUrls": [ { "title": "Demucs official documentation", "url": "https://github.com/facebookresearch/demucs" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Isolate voice from mixed audio", "stars": 5, "rationale": "The right specialised layer for vocal separation; compare stem quality on the actual recording.", "recommendation": "Use before transcription or mixing when music masks speech.", "domain": "Voice", "compareTask": "Isolate voice from mixed audio" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/demucs.jpg", "thumbnailEvidence": "Official diagram from the Meta (Facebook Research) Demucs GitHub README, retrieved 14 September 2026.", "comparisonNotes": "It does not transcribe or generate narration. Separation can introduce artefacts; it cannot recover every obscured word.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/facebookresearch/demucs" }, { "id": "elevenlabs-dubbing", "name": "ElevenLabs Dubbing", "domain": "Voice", "kind": "Hosted dubbing workflow", "compareTask": "Translate and dub a recording", "summary": "Translates recorded audio/video while retaining speaker characteristics.", "bestFor": "Choose managed dubbing when hosted processing and manual review are acceptable.", "notFor": "Not live interpretation; dubbing is separate from talking-face animation.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Hosted application/API", "inputs": [ "Audio or video", "Optional approved transcript" ], "outputs": [ "Translated dub" ], "limitations": [ "Documented dubbing is not realtime. Language support and limits can change." ], "sourceUrls": [ { "title": "ElevenLabs Dubbing official documentation", "url": "https://elevenlabs.io/docs/overview/capabilities/dubbing" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Translate and dub a recording", "stars": 4, "rationale": "A complete managed dubbing route; compare translation, timing and pronunciation before publication.", "recommendation": "Choose managed dubbing when hosted processing and manual review are acceptable.", "domain": "Voice", "compareTask": "Translate and dub a recording" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/elevenlabs-dubbing.jpg", "thumbnailEvidence": "Official share image (\"Dubbing V2 alpha\") from the ElevenLabs Dubbing Studio product page, retrieved 14 September 2026.", "comparisonNotes": "Not live interpretation; dubbing is separate from talking-face animation. Documented dubbing is not realtime. Language support and limits can change.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Voice" ], "thumbnailSource": "https://elevenlabs.io/dubbing-studio" }, { "id": "elevenlabs-agents", "name": "ElevenLabs Agents", "domain": "Voice", "kind": "Hosted agent platform", "compareTask": "Conversation services", "summary": "Platform for building conversational voice agents.", "bestFor": "Use for a hosted guide with configured knowledge and tools.", "notFor": "Narration or a visible talking portrait does not prove conversation works.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Hosted agent configuration and application integration", "inputs": [ "Live speech", "Knowledge and tools" ], "outputs": [ "Spoken responses", "Tool actions" ], "limitations": [ "Studio guide instances are not all proven live; test the deployed instance separately." ], "sourceUrls": [ { "title": "ElevenLabs Agents official documentation", "url": "https://elevenlabs.io/docs/overview/intro/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Conversation services", "stars": 4, "rationale": "A managed platform fits guide deployment, but the actual page still needs microphone, interruption and recovery tests.", "recommendation": "Use for a hosted guide with configured knowledge and tools.", "domain": "Voice", "compareTask": "Conversation services" } ], "relatedRecipePaths": [ "guide/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/elevenlabs-agents.jpg", "thumbnailEvidence": "Official share image from the ElevenLabs Agents app page, retrieved 14 September 2026.", "comparisonNotes": "Narration or a visible talking portrait does not prove conversation works. Studio guide instances are not all proven live; test the deployed instance separately.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Voice" ], "thumbnailSource": "https://elevenlabs.io/app/agents" }, { "id": "openai-realtime", "name": "OpenAI Realtime", "domain": "Voice", "kind": "Realtime model API", "compareTask": "Conversation services", "summary": "API for low-latency audio conversations and application tools.", "bestFor": "Choose for a custom conversational application where you control the integration.", "notFor": "Not a ready-made website guide or a dictation application.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Realtime API plus application transport", "inputs": [ "Live audio", "Instructions", "Tool definitions" ], "outputs": [ "Audio/text responses", "Tool calls" ], "limitations": [ "Session behaviour, voices, access and supported models must be checked for the selected API release." ], "sourceUrls": [ { "title": "OpenAI Realtime official documentation", "url": "https://platform.openai.com/docs/api-reference/realtime" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Conversation services", "stars": 4, "rationale": "Strong fit for custom integration; requires more application work than a hosted agent builder.", "recommendation": "Choose for a custom conversational application where you control the integration.", "domain": "Voice", "compareTask": "Conversation services" } ], "relatedRecipePaths": [ "guide/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/openai-realtime.jpg", "thumbnailEvidence": "GitHub repository social preview card for OpenAI's official openai-realtime-console reference app, retrieved 14 September 2026.", "comparisonNotes": "Not a ready-made website guide or a dictation application. Session behaviour, voices, access and supported models must be checked for the selected API release.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/openai/openai-realtime-console" }, { "id": "gemini-live", "name": "Gemini Live", "domain": "Voice", "kind": "Realtime multimodal API", "compareTask": "Conversation services", "summary": "Realtime API for conversational audio and multimodal sessions.", "bestFor": "Choose when live voice must interact with visual context in a Google-based integration.", "notFor": "Not the same capability as ordinary text-only Gemini calls.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Live API and application integration", "inputs": [ "Streaming audio", "Optional visual context" ], "outputs": [ "Audio/text responses" ], "limitations": [ "Verify model availability, session limits and interruption in the actual application." ], "sourceUrls": [ { "title": "Gemini Live official documentation", "url": "https://ai.google.dev/gemini-api/docs/live-api" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Conversation services", "stars": 4, "rationale": "A differentiated route for multimodal interaction; not an acoustic-quality ranking against other voice APIs.", "recommendation": "Choose when live voice must interact with visual context in a Google-based integration.", "domain": "Voice", "compareTask": "Conversation services" } ], "relatedRecipePaths": [ "guide/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/gemini-live.jpg", "thumbnailEvidence": "Official share thumbnail from the Gemini app homepage (gemini.google.com), retrieved 14 September 2026.", "comparisonNotes": "Not the same capability as ordinary text-only Gemini calls. Verify model availability, session limits and interruption in the actual application.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Voice" ], "thumbnailSource": "https://gemini.google.com/" }, { "id": "grok-voice", "name": "Grok Voice Agent API", "domain": "Voice", "kind": "Realtime model API", "compareTask": "Conversation services", "summary": "Realtime voice interface with configurable tools.", "bestFor": "Evaluate when an xAI-based voice integration is specifically useful.", "notFor": "Grok’s speaking-video generation is a separate product and workflow.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Voice Agent API", "inputs": [ "Live audio", "Tools" ], "outputs": [ "Spoken responses", "Tool calls" ], "limitations": [ "Do not infer API access from a consumer subscription." ], "sourceUrls": [ { "title": "Grok Voice Agent API official documentation", "url": "https://docs.x.ai/developers/model-capabilities/audio/voice-agent" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Conversation services", "stars": 3, "rationale": "A relevant alternative; no current Studio guide integration was tested here.", "recommendation": "Evaluate when an xAI-based voice integration is specifically useful.", "domain": "Voice", "compareTask": "Conversation services" } ], "relatedRecipePaths": [ "guide/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/grok-voice.jpg", "thumbnailEvidence": "Official share image from xAI's Grok Voice Agent API documentation page, retrieved 14 September 2026.", "comparisonNotes": "Grok’s speaking-video generation is a separate product and workflow. Do not infer API access from a consumer subscription.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Voice" ], "thumbnailSource": "https://docs.x.ai/developers/model-capabilities/audio/voice-agent" }, { "id": "pipecat", "name": "Pipecat", "domain": "Voice", "kind": "Voice-agent framework", "compareTask": "Build your own conversation stack", "summary": "Framework connecting realtime transports and voice/AI services.", "bestFor": "Use when you need control over model providers and the conversation pipeline.", "notFor": "Not a competing voice model or a hosted voice by itself.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application code with chosen providers", "inputs": [ "Audio transport", "ASR/LLM/TTS components" ], "outputs": [ "Working voice-agent pipeline" ], "limitations": [ "Each provider adds its own cost, privacy, latency and availability constraints." ], "sourceUrls": [ { "title": "Pipecat official documentation", "url": "https://docs.pipecat.ai/overview/introduction" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Build your own conversation stack", "stars": 5, "rationale": "Best fit within this set for composing a provider-selectable pipeline; engineering and deployment are still required.", "recommendation": "Use when you need control over model providers and the conversation pipeline.", "domain": "Voice", "compareTask": "Build your own conversation stack" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/pipecat.jpg", "thumbnailEvidence": "Headless capture of the official Pipecat homepage, retrieved 14 September 2026.", "comparisonNotes": "Not a competing voice model or a hosted voice by itself. Each provider adds its own cost, privacy, latency and availability constraints.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Voice" ], "thumbnailSource": "https://pipecat.ai/" }, { "id": "hf-speech-to-speech", "name": "Hugging Face speech-to-speech", "domain": "Voice", "kind": "Reference application stack", "compareTask": "Build your own conversation stack", "summary": "Open-source assembly of components for voice agents.", "bestFor": "Use as a reference or starting implementation for an open-model pipeline.", "notFor": "Not a single end-to-end model or an already-deployed service.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Reference code and selected models", "inputs": [ "Audio", "Models and runtime" ], "outputs": [ "Speech-to-speech pipeline" ], "limitations": [ "Hardware needs depend on the selected components." ], "sourceUrls": [ { "title": "Hugging Face speech-to-speech official documentation", "url": "https://github.com/huggingface/speech-to-speech" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Build your own conversation stack", "stars": 3, "rationale": "Useful for learning and prototyping; adaptation and deployment checks remain.", "recommendation": "Use as a reference or starting implementation for an open-model pipeline.", "domain": "Voice", "compareTask": "Build your own conversation stack" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/hf-speech-to-speech.jpg", "thumbnailEvidence": "Endpoint-swap demo GIF frame from the official Hugging Face speech-to-speech GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Not a single end-to-end model or an already-deployed service. Hardware needs depend on the selected components.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/huggingface/speech-to-speech" }, { "id": "moshi", "name": "Moshi", "domain": "Voice", "kind": "Dialogue model and framework", "compareTask": "Open dialogue models", "summary": "Speech-text model and full-duplex spoken-dialogue research framework.", "bestFor": "Evaluate when research-level control over full-duplex dialogue is the objective.", "notFor": "Not the simplest production route for a website guide.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Self-hosted model/runtime", "inputs": [ "Live audio" ], "outputs": [ "Spoken dialogue" ], "limitations": [ "Validate languages, hardware and behaviour on the intended conversation." ], "sourceUrls": [ { "title": "Moshi official documentation", "url": "https://github.com/kyutai-labs/moshi" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Open dialogue models", "stars": 3, "rationale": "Distinct from a modular ASR-LLM-TTS stack; useful to evaluate, with substantial deployment work.", "recommendation": "Evaluate when research-level control over full-duplex dialogue is the objective.", "domain": "Voice", "compareTask": "Open dialogue models" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/moshi.jpg", "thumbnailEvidence": "Official diagram from the Kyutai Labs Moshi GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Not the simplest production route for a website guide. Validate languages, hardware and behaviour on the intended conversation.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/kyutai-labs/moshi" }, { "id": "nemotron-voicechat", "name": "Nemotron VoiceChat", "domain": "Voice", "kind": "Dialogue model", "compareTask": "Open dialogue models", "summary": "NVIDIA real-time conversational speech model.", "bestFor": "Keep as a candidate for a self-hosted dialogue evaluation.", "notFor": "Do not mark available on the current machine without a deployment check.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Model/runtime deployment", "inputs": [ "Audio stream" ], "outputs": [ "Conversational audio" ], "limitations": [ "Verify the exact model card, hardware and supported deployment." ], "sourceUrls": [ { "title": "Nemotron VoiceChat official documentation", "url": "https://build.nvidia.com/nvidia/nemotron-voicechat/experience" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no listening benchmark", "ratings": [ { "task": "Open dialogue models", "stars": null, "rationale": "Official model identified, but there is no local integration or comparative test in this review.", "recommendation": "Keep as a candidate for a self-hosted dialogue evaluation.", "domain": "Voice", "compareTask": "Open dialogue models" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/nemotron-voicechat.jpg", "thumbnailEvidence": "Official share image from NVIDIA's Nemotron VoiceChat page on build.nvidia.com, retrieved 14 September 2026.", "comparisonNotes": "Do not mark available on the current machine without a deployment check. Verify the exact model card, hardware and supported deployment.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://build.nvidia.com/nvidia/nemotron-voicechat/experience" }, { "id": "kokoro", "name": "Kokoro", "domain": "Voice", "kind": "Speech engine or client", "compareTask": "Choose an underlying speech engine", "summary": "Compact text-to-speech model with supplied voices.", "bestFor": "Choose for lightweight speech synthesis using available voices.", "notFor": "Not a complete voice studio or conversational system.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Engine/client, often exposed through voice-pro", "inputs": [ "Text", "Reference audio where supported" ], "outputs": [ "Speech audio" ], "limitations": [ "Not the choice when reproducing a particular reference speaker is mandatory; check language and voice support." ], "sourceUrls": [ { "title": "Kokoro official model/project documentation", "url": "https://huggingface.co/hexgrad/Kokoro-82M" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official model/project documentation reviewed; no same-script listening test", "ratings": [ { "task": "Choose an underlying speech engine", "stars": 4, "rationale": "Editorial task fit based on the documented role and integration requirements. Not the choice when reproducing a particular reference speaker is mandatory; check language and voice support.", "recommendation": "Choose for lightweight speech synthesis using available voices.", "domain": "Voice", "compareTask": "Choose an underlying speech engine" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/kokoro.jpg", "thumbnailEvidence": "Official Hugging Face social-thumbnail image for the Kokoro-82M model card, retrieved 14 September 2026.", "comparisonNotes": "Not the choice when reproducing a particular reference speaker is mandatory; check language and voice support.", "costModel": "Local inference compute; check distribution terms.", "dataLocation": "Local when self-hosted.", "languageSupport": "Use the voices and languages published with the selected release.", "domains": [ "Voice" ], "thumbnailSource": "https://huggingface.co/hexgrad/Kokoro-82M" }, { "id": "edge-tts", "name": "Edge-TTS", "domain": "Voice", "kind": "Speech engine or client", "compareTask": "Choose an underlying speech engine", "summary": "Python client for Microsoft Edge’s online text-to-speech service.", "bestFor": "Use for a simple online speech-generation route when that service dependency is acceptable.", "notFor": "Not a complete voice studio or conversational system.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Engine/client, often exposed through voice-pro", "inputs": [ "Text", "Reference audio where supported" ], "outputs": [ "Speech audio" ], "limitations": [ "This is an online client, not a local model. Do not infer a production service guarantee from a working client." ], "sourceUrls": [ { "title": "Edge-TTS official model/project documentation", "url": "https://github.com/rany2/edge-tts" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official model/project documentation reviewed; no same-script listening test", "ratings": [ { "task": "Choose an underlying speech engine", "stars": 3, "rationale": "Editorial task fit based on the documented role and integration requirements. This is an online client, not a local model. Do not infer a production service guarantee from a working client.", "recommendation": "Use for a simple online speech-generation route when that service dependency is acceptable.", "domain": "Voice", "compareTask": "Choose an underlying speech engine" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/edge-tts.jpg", "thumbnailEvidence": "GitHub repository social preview card for rany2/edge-tts, retrieved 14 September 2026.", "comparisonNotes": "This is an online client, not a local model. Do not infer a production service guarantee from a working client.", "costModel": "Client uses an online service; check applicable terms.", "dataLocation": "Online service.", "languageSupport": "Choose a voice exposed by the service; this is not arbitrary speaker cloning.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/rany2/edge-tts" }, { "id": "f5-tts", "name": "F5-TTS", "domain": "Voice", "kind": "Speech engine or client", "compareTask": "Choose an underlying speech engine", "summary": "Flow-matching text-to-speech system for reference-voice generation.", "bestFor": "Evaluate as a self-hosted reference-voice engine inside a supported application.", "notFor": "Not a complete voice studio or conversational system.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Engine/client, often exposed through voice-pro", "inputs": [ "Text", "Reference audio where supported" ], "outputs": [ "Speech audio" ], "limitations": [ "The project distinguishes MIT code from non-commercial pretrained weights. Verify the intended use against the selected model terms." ], "sourceUrls": [ { "title": "F5-TTS official model/project documentation", "url": "https://github.com/SWivid/F5-TTS" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official model/project documentation reviewed; no same-script listening test", "ratings": [ { "task": "Choose an underlying speech engine", "stars": 3, "rationale": "Editorial task fit based on the documented role and integration requirements. Verify model weights, licence and reference preparation. Application code and pretrained weights may have different terms.", "recommendation": "Evaluate as a self-hosted reference-voice engine inside a supported application.", "domain": "Voice", "compareTask": "Choose an underlying speech engine" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/f5-tts.jpg", "thumbnailEvidence": "GitHub repository social preview card for SWivid/F5-TTS, retrieved 14 September 2026.", "comparisonNotes": "Verify model weights, licence and reference preparation. Application code and pretrained weights may have different terms.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/SWivid/F5-TTS" }, { "id": "cosyvoice", "name": "CosyVoice", "domain": "Voice", "kind": "Speech engine or client", "compareTask": "Choose an underlying speech engine", "summary": "Multilingual speech-generation family with inference and deployment tooling.", "bestFor": "Choose when you need a controllable multilingual model route in an application you can operate.", "notFor": "Not a complete voice studio or conversational system.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Engine/client, often exposed through voice-pro", "inputs": [ "Text", "Reference audio where supported" ], "outputs": [ "Speech audio" ], "limitations": [ "Pin the version and test the intended language, reference voice and streaming mode." ], "sourceUrls": [ { "title": "CosyVoice official model/project documentation", "url": "https://github.com/QwenAudio/CosyVoice" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official model/project documentation reviewed; no same-script listening test", "ratings": [ { "task": "Choose an underlying speech engine", "stars": 4, "rationale": "Editorial task fit based on the documented role and integration requirements. Pin the version and test the intended language, reference voice and streaming mode.", "recommendation": "Choose when you need a controllable multilingual model route in an application you can operate.", "domain": "Voice", "compareTask": "Choose an underlying speech engine" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/cosyvoice.jpg", "thumbnailEvidence": "Official image from the QwenAudio CosyVoice GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Pin the version and test the intended language, reference voice and streaming mode.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/QwenAudio/CosyVoice" }, { "id": "qwen3-tts", "name": "Qwen3-TTS", "domain": "Voice", "kind": "Speech-generation engine", "compareTask": "Choose an underlying speech engine", "summary": "Speech generation family with separate voice-design, custom-voice and cloning models.", "bestFor": "Choose when you need instruction-led voice design or a supported cloning route.", "notFor": "Do not treat a model listing as installed and ready.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Model runtime or compatible application", "inputs": [ "Text", "Reference audio where supported" ], "outputs": [ "Speech audio" ], "limitations": [ "VoiceDesign, CustomVoice and Base are different models. Pick the correct one; advertised cloning input length is not an output-quality guarantee." ], "sourceUrls": [ { "title": "Qwen3-TTS official model/project documentation", "url": "https://github.com/QwenLM/Qwen3-TTS" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official model/project documentation reviewed; no same-script listening test", "ratings": [ { "task": "Choose an underlying speech engine", "stars": 4, "rationale": "Editorial task fit based on the documented role and integration requirements. VoiceDesign, CustomVoice and Base are different models. Pick the correct one; advertised cloning input length is not an output-quality guarantee.", "recommendation": "Choose when you need instruction-led voice design or a supported cloning route.", "domain": "Voice", "compareTask": "Choose an underlying speech engine" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/qwen3-tts.jpg", "thumbnailEvidence": "Official introduction image from the Qwen3-TTS GitHub README (hosted on Alibaba's own CDN), retrieved 14 September 2026.", "comparisonNotes": "VoiceDesign, CustomVoice and Base are different models. Pick the correct one; advertised cloning input length is not an output-quality guarantee.", "costModel": "Model compute; deployment or hosted-service charges.", "dataLocation": "Local/self-hosted, or selected cloud provider.", "languageSupport": "Official series includes English, French, Spanish and other named languages; model variant matters.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/QwenLM/Qwen3-TTS" }, { "id": "chatterbox", "name": "Chatterbox", "domain": "Voice", "kind": "Speech-generation engine", "compareTask": "Choose an underlying speech engine", "summary": "Open speech-generation model family with voice-reference and expressiveness controls.", "bestFor": "Choose for reference-voice speech generation when its supported version and language fit.", "notFor": "Do not treat a model listing as installed and ready.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Model runtime or compatible application", "inputs": [ "Text", "Reference audio where supported" ], "outputs": [ "Speech audio" ], "limitations": [ "Versions and runtimes vary; compare pronunciation and delivery by ear before production." ], "sourceUrls": [ { "title": "Chatterbox official model/project documentation", "url": "https://github.com/resemble-ai/chatterbox" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official model/project documentation reviewed; no same-script listening test", "ratings": [ { "task": "Choose an underlying speech engine", "stars": 4, "rationale": "Editorial task fit based on the documented role and integration requirements. Versions and runtimes vary; compare pronunciation and delivery by ear before production.", "recommendation": "Choose for reference-voice speech generation when its supported version and language fit.", "domain": "Voice", "compareTask": "Choose an underlying speech engine" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/chatterbox.jpg", "thumbnailEvidence": "Official image from the Resemble AI Chatterbox GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Versions and runtimes vary; compare pronunciation and delivery by ear before production.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/resemble-ai/chatterbox" }, { "id": "indextts2", "name": "IndexTTS 2 / 2.5", "domain": "Voice", "kind": "Speech-generation engine", "compareTask": "Choose an underlying speech engine", "summary": "Controllable zero-shot speech-generation system.", "bestFor": "Evaluate when precise control over the generated speech is needed.", "notFor": "Do not treat a model listing as installed and ready.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Model runtime or compatible application", "inputs": [ "Text", "Reference audio where supported" ], "outputs": [ "Speech audio" ], "limitations": [ "Do not transfer capabilities between versions: the 2.0 release notes say precise duration control was not enabled; 2.5 documents a speaking-speed duration factor. Check supported languages." ], "sourceUrls": [ { "title": "IndexTTS2 official model/project documentation", "url": "https://github.com/index-tts/index-tts" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official model/project documentation reviewed; no same-script listening test", "ratings": [ { "task": "Choose an underlying speech engine", "stars": 3, "rationale": "Editorial task fit based on the documented role and integration requirements. Check the released model and supported controls. A documented feature is not a verified local integration.", "recommendation": "Evaluate when precise control over the generated speech is needed.", "domain": "Voice", "compareTask": "Choose an underlying speech engine" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/indextts2.jpg", "thumbnailEvidence": "Official video-cover image from the IndexTTS2/2.5 GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Do not transfer capabilities between versions: the 2.0 release notes say precise duration control was not enabled; 2.5 documents a speaking-speed duration factor. Check supported languages.", "costModel": "Model compute; check the selected release and weights.", "dataLocation": "Self-hosted/runtime dependent.", "languageSupport": "IndexTTS 2.5 documents Chinese, English, Japanese, Spanish and Arabic; do not assume French support.", "domains": [ "Voice" ], "thumbnailSource": "https://github.com/index-tts/index-tts" }, { "id": "auk-flash", "name": "AuK-Flash", "domain": "Voice", "kind": "Speech-generation engine", "compareTask": "Choose an underlying speech engine", "summary": "Tencent's AuK (announced as \"Nano Banana for audio\"), here in its distilled four-step version: speech from a reference voice or a written voice description, plus editing, cleaning and separation of existing speech through written instructions.", "bestFor": "Repairing an existing voice track by instruction (swap one mispronounced word, change emotion, remove breaths or noise) and making extra takes of a speaker's voice from a short reference, in non-commercial work.", "notFor": "Not for commercial use while it depends on the Qwen research encoder, not a streaming voice, and French is still untested.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Self-hosted Python runtime on a CUDA GPU, or the MLX branch on Apple Silicon; Gradio, ComfyUI and SGLang-Omni integrations", "inputs": [ "Written instruction with the text to speak", "Reference audio to clone, or source audio to edit" ], "outputs": [ "Speech audio at 24 kHz", "Edited, cleaned or separated recording" ], "limitations": [ "The model card, instruction templates and published evaluations cover English and Chinese only. One Studio test on 27 September 2026 produced correct Spanish (see comparison notes); French is untested.", "It needs the Qwen2.5-Omni-3B encoder, whose Qwen Research License allows non-commercial use only. The AuK weights themselves are MIT.", "Output length is fixed in advance: pass a target duration, or let the optional prompt enhancer estimate it through a separate language model. No streaming. The demo caps a take at 30 seconds.", "Launch-day peak GPU memory was about 25 GB; a 16 September update cut the encoder by about 7.5 GB and added CPU offload. Apple Silicon runs from a separate MLX branch; audio.cpp added offline GGUF runs on 20 September.", "The free demo allows about one 30-second take a day without a Hugging Face token (measured 27 September 2026)." ], "sourceUrls": [ { "title": "AuK-Flash official model card", "url": "https://huggingface.co/tencent/AuK-Flash" }, { "title": "AuK official code repository", "url": "https://github.com/Tencent-Hunyuan/AuK" }, { "title": "AuK licence file (MIT)", "url": "https://huggingface.co/tencent/AuK-Flash/blob/main/LICENSE" }, { "title": "AuK technical report", "url": "https://arxiv.org/abs/2609.08936" }, { "title": "AuK Apple Silicon (MLX) branch", "url": "https://github.com/Tencent-Hunyuan/AuK/tree/feat/mlx-apple-silicon" }, { "title": "Qwen2.5-Omni-3B encoder licence (Qwen Research License)", "url": "https://huggingface.co/Qwen/Qwen2.5-Omni-3B/blob/main/LICENSE" }, { "title": "AuK public demo space", "url": "https://huggingface.co/spaces/tencent/AuK" } ], "sourceChecked": "2026-09-27", "evidenceLevel": "Official model card, repository, licence files and technical report reviewed; one Spanish zero-shot take measured by Studio on the public demo (27 September 2026), not yet judged by ear", "ratings": [ { "task": "Choose an underlying speech engine", "stars": 2, "rationale": "Editorial task fit. Spanish now works in one Studio test, but its required encoder is non-commercial and it needs a large GPU or the Apple Silicon branch, so it stays below IndexTTS 2 / 2.5 (three stars) and Qwen3-TTS (four stars) as a general engine. Its instruction editing of existing speech goes further than any other engine in this task.", "recommendation": "Use the public demo to repair or extend an existing voice track (one word, one emotion, extra takes) and judge it by ear. Do not build a commercial product on it while the Qwen encoder licence stays research-only.", "domain": "Voice", "compareTask": "Choose an underlying speech engine" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/auk-flash.jpg", "thumbnailEvidence": "Official Hugging Face social-thumbnail image for Tencent's AuK-Flash model card, retrieved 14 September 2026.", "comparisonNotes": "The editing range is what sets it apart: replace spoken words, change emotion or timbre, reduce an accent, add or remove breaths and laughs, convert whisper, denoise and keep one speaker, each through a written instruction. Studio test, 27 September 2026, AuK Base on the public demo: a 15-second Spanish reference from an existing Studio speaker and a 20-word Spanish sentence. Speech recognition returned every word exactly; median pitch 204 Hz against 198 Hz for the reference (+6 Hz, half the 12 Hz drift measured between takes of the engine that invented that voice); loudness -11.5 against -12.1 LUFS; spectral centroid 1,075 Hz against 941 Hz, so slightly brighter. Timbre match still needs a listening check. The authors' report compares it with Qwen3-TTS and VoxCPM2 on English and Chinese test sets only; Studio has not verified those results.", "costModel": "MIT weights, self-hosted GPU or Mac compute. Commercial use also needs a separate licence for the Qwen encoder.", "dataLocation": "Self-hosted if run locally; the public demo runs on Hugging Face.", "languageSupport": "Model card lists Chinese and English; instruction templates and evaluations are English and Chinese only. Spanish worked in one Studio test (27 September 2026). French untested.", "domains": [ "Voice" ], "thumbnailSource": "https://huggingface.co/tencent/AuK-Flash" }, { "id": "meta-voice-transcribe", "name": "Meta Voice Transcribe", "domain": "Voice", "kind": "Unresolved product label", "compareTask": "Unresolved identities", "summary": "A name in the existing material whose exact product/API identity needs resolution.", "bestFor": "Keep for follow-up identification.", "notFor": "Do not select for a real workflow yet.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Unverified", "inputs": [ "Unverified" ], "outputs": [ "Unverified" ], "limitations": [ "No current primary source identified." ], "sourceUrls": [], "sourceChecked": null, "evidenceLevel": "Unresolved identity", "ratings": [ { "task": "Unresolved identities", "stars": null, "rationale": "An ambiguous name cannot support a fair comparison.", "recommendation": "Keep for follow-up identification.", "domain": "Voice", "compareTask": "Unresolved identities" } ], "relatedRecipePaths": [ "voice/index.html" ], "thumbnailCandidate": "catalogue/covers/meta-voice-transcribe.svg", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "comparisonNotes": "Do not select for a real workflow yet. No current primary source identified.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Voice" ] }, { "id": "gemini-image", "name": "Gemini image / Nano Banana", "domain": "Images", "kind": "Image engine", "summary": "Generates and edits raster images through Google image-capable models.", "bestFor": "Conversational illustration and reference-based edits", "notFor": "Editable SVG or factual diagrams without checking", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Gemini API or Google AI Studio", "inputs": [ "Prompt", "Reference images" ], "outputs": [ "Raster image" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Gemini image / Nano Banana official source", "url": "https://ai.google.dev/gemini-api/docs/image-generation" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Conversational illustration and reference-based edits", "stars": 4, "rationale": "Native image input/output supports iterative illustration; no evidence here that it beats another engine.", "recommendation": "Use for this task", "compareTask": "Generate and edit illustrations", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/gemini-image.jpg", "comparisonNotes": "Google AI Studio is an interface, Nano Banana a model nickname. Select an exact model ID per job. Existing page claims of fractions of a cent and live access are not carried forward.", "thumbnailEvidence": "Official share image from the Gemini API image-generation documentation page, retrieved 14 September 2026.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images" ], "thumbnailSource": "https://ai.google.dev/gemini-api/docs/image-generation" }, { "id": "gpt-image", "name": "GPT Image 2", "domain": "Images", "kind": "Image engine", "summary": "Generates and edits raster images with text instructions and image inputs.", "bestFor": "Detailed prompted illustration and editing", "notFor": "Editable design components or guaranteed exact labels", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "OpenAI Images API; available host image-generation tool where exposed", "inputs": [ "Prompt", "Reference images" ], "outputs": [ "Raster image" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "GPT Image 2 official source", "url": "https://developers.openai.com/api/docs/models/gpt-image-2" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Detailed prompted illustration and editing", "stars": 4, "rationale": "Official generation and editing support is a strong task fit; no comparative quality winner is established.", "recommendation": "Use for this task", "compareTask": "Generate and edit illustrations", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/covers/gpt-image.svg", "comparisonNotes": "Do not assume API access, an app subscription and a third-party wrapper are interchangeable entitlements.", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images" ] }, { "id": "flux2-local", "name": "FLUX.2 dev on RunPod", "domain": "Images", "kind": "Self-hosted image engine", "summary": "Open-weight FLUX image generation operated through the existing GPU workflow.", "bestFor": "Repeatable batches through an owned workflow", "notFor": "Zero-maintenance service or guaranteed cheaper output", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Existing video toolkit GPU wrapper; verify deployed model and endpoint", "inputs": [ "Prompt", "Optional references if deployed version supports them" ], "outputs": [ "Raster images" ], "limitations": [ "Compute, deployed model variant and licence must be checked for the selected job.", "Current health and throughput were not tested." ], "sourceUrls": [ { "title": "FLUX.2 dev on RunPod official source", "url": "https://docs.bfl.ai/flux_2/flux2_overview" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Repeatable batches through an owned workflow", "stars": 4, "rationale": "Recorded batch workflow is useful; suitability is operational, not evidence of superior image quality.", "recommendation": "Use for this task", "compareTask": "Generate and edit illustrations", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/flux2-local.jpg", "comparisonNotes": "Keep dev separate from BFL hosted pro/flex/max. Do not call them the same model or carry forward old per-image cost.", "thumbnailEvidence": "Official Hugging Face social-thumbnail image for the Black Forest Labs FLUX.2-dev (open-weights) model card, retrieved 14 September 2026.", "costModel": "Rented/local GPU time and model/weight terms.", "dataLocation": "Self-hosted on selected hardware; cloud if that hardware is rented.", "languageSupport": "Prompt/model coverage must be tested.", "domains": [ "Images" ], "thumbnailSource": "https://huggingface.co/black-forest-labs/FLUX.2-dev" }, { "id": "flux2-hosted", "name": "FLUX.2 hosted models", "domain": "Images", "kind": "Hosted image engine", "summary": "BFL endpoints generate images and perform multi-reference edits.", "bestFor": "Hosted reference-based generation and editing", "notFor": "Native editable SVG", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "BFL API; select exact variant", "inputs": [ "Prompt", "Reference images" ], "outputs": [ "Raster images" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "FLUX.2 hosted models official source", "url": "https://docs.bfl.ai/flux_2/flux2_image_editing" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Hosted reference-based generation and editing", "stars": 4, "rationale": "Hosted API and documented reference editing fit automation; no benchmark comparison with local dev was run.", "recommendation": "Use for this task", "compareTask": "Generate and edit illustrations", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/flux2-hosted.jpg", "comparisonNotes": "Hosted variant and self-hosted dev are separate choices. Provider marketing about speed or quality is not a measured Studio rating.", "thumbnailEvidence": "Headless capture of the official Black Forest Labs homepage, retrieved 14 September 2026. Shows BFL's current flagship model lineup, the vendor behind FLUX.2's hosted API.", "costModel": "Hosted inference charges.", "dataLocation": "Provider-hosted image processing.", "languageSupport": "Prompt/model coverage must be tested.", "domains": [ "Images" ], "thumbnailSource": "https://bfl.ai/" }, { "id": "recraft", "name": "Recraft", "domain": "Images", "kind": "Raster/vector image engine", "summary": "Provides generation, vector generation, vectorization and editing endpoints.", "bestFor": "Generated vector illustrations and icons", "notFor": "Assuming a generated diagram is logically correct", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Recraft API or editor; choose vector model for vector output", "inputs": [ "Prompt", "Reference image", "Mask for relevant editing endpoint" ], "outputs": [ "Raster images", "SVG vectors" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Recraft official source", "url": "https://www.recraft.ai/docs/api-reference/endpoints" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Generated vector illustrations and icons", "stars": 5, "rationale": "Documented vector output directly matches editable illustration; five stars mean task suitability, not best artwork.", "recommendation": "Use for this task", "compareTask": "Create editable vector artwork", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/recraft.jpg", "comparisonNotes": "Separate native vector generation from raster-to-vector conversion. Generated SVG still needs geometry and text review; it is not uniquely the only tool that can produce vectors.", "thumbnailEvidence": "Headless capture of the official Recraft homepage, retrieved 14 September 2026.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images" ], "thumbnailSource": "https://www.recraft.ai/" }, { "id": "ideogram", "name": "Ideogram", "domain": "Images", "kind": "Image engine", "summary": "Image generation and editing with typography-oriented features and structured prompts.", "bestFor": "Posters, branding studies and compositions with visible text", "notFor": "Accessible live UI or final institutional labels inside bitmaps", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Ideogram API or application", "inputs": [ "Text prompt", "Structured prompt", "Image for remix/edit" ], "outputs": [ "Raster images" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Ideogram official source", "url": "https://developer.ideogram.ai/ideogram-api/api-overview" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Posters, branding studies and compositions with visible text", "stars": 4, "rationale": "Typography is a documented product focus; treat text as artwork and proofread it.", "recommendation": "Use for this task", "compareTask": "Compose artwork with text", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/ideogram.jpg", "comparisonNotes": "Do not repeat a leaderboard position as Studio evidence. Text embedded in a picture does not become a usable interface.", "thumbnailEvidence": "Official share image from the Ideogram homepage, retrieved 14 September 2026.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images" ], "thumbnailSource": "https://ideogram.ai/" }, { "id": "seedream", "name": "Seedream", "domain": "Images", "kind": "Image engine", "summary": "ByteDance image-generation family exposed through BytePlus and selected providers.", "bestFor": "Alternative image generation to evaluate on the same brief", "notFor": "An assumed quality improvement over existing engines", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "BytePlus ModelArk or a verified provider", "inputs": [ "Prompt", "Reference images where model supports them" ], "outputs": [ "Raster images" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Seedream official source", "url": "https://www.byteplus.com/en/activity/free" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official product listing checked; detailed API page rendered only its shell.", "ratings": [ { "task": "Alternative image generation to evaluate on the same brief", "stars": null, "rationale": "Official source confirms generation/editing offering; insufficient comparative or local evidence to rate.", "recommendation": "Not rated; verify before choosing", "compareTask": "Generate and edit illustrations", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/seedream.jpg", "comparisonNotes": "Keep Seedream (image) distinct from Seedance (video). Exact model/version and available editing controls require confirmation.", "thumbnailEvidence": "Official BytePlus share image from the Seedream source page, retrieved 14 September 2026.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images" ], "thumbnailSource": "https://www.byteplus.com/en/activity/free" }, { "id": "midjourney", "name": "Midjourney", "domain": "Images", "kind": "Image engine and editor", "summary": "Generates visual interpretations from text and image references; web editor supports targeted changes.", "bestFor": "Human-led visual exploration", "notFor": "Assumed unattended API automation", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Web application; image prompts and editor", "inputs": [ "Prompt", "Image reference" ], "outputs": [ "Raster images" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Midjourney official source", "url": "https://docs.midjourney.com/hc/en-us/articles/32040250122381-Image-Prompts" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Human-led visual exploration", "stars": 4, "rationale": "Reference-based exploration is documented; use when a person directs iterations.", "recommendation": "Use for this task", "compareTask": "Explore a visual direction", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/covers/midjourney.svg", "comparisonNotes": "Existing rejection concerned automation, not poor image quality. Current broad public API availability was not established; avoid an absolute no-API claim.", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images" ] }, { "id": "krea", "name": "Krea", "domain": "Images", "kind": "Image platform and model host", "summary": "Image creation and enhancement platform with a documented developer API.", "bestFor": [ "Exploring image generation and enhancement workflows", "Choose for interactive exploration when its available models and controls fit the task." ], "notFor": [ "Assuming its old free quota or web-only limits still apply", "Krea output depends on the selected model; it is not one video engine." ], "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Krea application or API; confirm chosen model", "inputs": [ "Prompt", "Images", "Supported references" ], "outputs": [ "Images", "Generated or edited media" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking.", "API/interface features and provider access may differ." ], "sourceUrls": [ { "title": "Krea official source", "url": "https://www.krea.ai/docs/" }, { "title": "Krea official source", "url": "https://www.krea.ai/docs/user-guide/features/video" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Exploring image generation and enhancement workflows", "stars": null, "rationale": "Official documentation confirms platform/API; current model-specific benefits and local experience are insufficient to rank.", "recommendation": "Not rated; verify before choosing", "compareTask": "Explore a visual direction", "domain": "Images" }, { "task": "Video creation workspaces", "stars": 3, "rationale": "Useful interface alternative, but no current Studio production test establishes it as the default.", "recommendation": "Choose for interactive exploration when its available models and controls fit the task.", "domain": "Images", "compareTask": "Video creation workspaces" } ], "relatedRecipePaths": [ "diagramas/index.html", "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/krea.jpg", "comparisonNotes": "Old Studio table says no API; current official docs include a developer API. Distinguish Krea platform from Krea-branded models.", "thumbnailEvidence": "Headless capture of the official Krea homepage, retrieved 14 September 2026.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images", "Video" ], "thumbnailSource": "https://www.krea.ai/" }, { "id": "higgsfield", "name": "Higgsfield", "domain": "Images", "kind": "Multi-model platform", "summary": "Routes image and video jobs through an agent-accessible platform.", "bestFor": [ "Trying several available models through one interface", "Use when trying several hosted models and camera/reference controls in one workflow." ], "notFor": [ "Treating wrapper name as image-quality evidence", "The workspace is not one model; compare the selected engine explicitly." ], "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Official MCP or CLI, signed-in account", "inputs": [ "Prompt", "Reference assets", "Reference images/video/audio as supported" ], "outputs": [ "Images", "Videos depending on model", "Generated or edited media" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking.", "Verify the chosen connection and entitlements. Do not assume one subscription makes every model unlimited." ], "sourceUrls": [ { "title": "Higgsfield for images official source", "url": "https://higgsfield.ai/mcp" }, { "title": "Higgsfield official source", "url": "https://github.com/higgsfield-ai/cli" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Trying several available models through one interface", "stars": 3, "rationale": "Convenient model access; suitability depends on selected engine and account access.", "recommendation": "Consider for this task", "compareTask": "Operate multiple image engines", "domain": "Images" }, { "task": "Video creation workspaces", "stars": 4, "rationale": "Useful shared creative interface. Official CLI now exists, so a historic MCP blockage is not a blanket capability verdict.", "recommendation": "Use when trying several hosted models and camera/reference controls in one workflow.", "domain": "Images", "compareTask": "Video creation workspaces" } ], "relatedRecipePaths": [ "diagramas/index.html", "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/higgsfield.jpg", "comparisonNotes": "Store the underlying model in every result. A platform comparison cannot substitute for an engine comparison.", "thumbnailEvidence": "Capture of the Higgsfield homepage, 14 September 2026.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images", "Video" ], "thumbnailSource": "https://higgsfield.ai/" }, { "id": "heliosgen", "name": "HeliosGen", "domain": "Images", "kind": "Workflow wrapper", "summary": "Visual canvas for chaining prompts, reference images, models and automation.", "bestFor": "Designing multi-step generation workflows", "notFor": "Image generation without a configured backend", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Desktop application and configured provider route", "inputs": [ "Prompts", "Reference images", "Workflow" ], "outputs": [ "Provider-generated images and video" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "HeliosGen official source", "url": "https://github.com/SegFault42/HeliosGen" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Designing multi-step generation workflows", "stars": null, "rationale": "Repository describes orchestration; current local installation and provider route untested.", "recommendation": "Not rated; verify before choosing", "compareTask": "Operate multiple image engines", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/heliosgen.jpg", "comparisonNotes": "Do not equate self-hosted application with local model inference or free generation.", "thumbnailEvidence": "GitHub repository social preview card for SegFault42/HeliosGen, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://github.com/SegFault42/HeliosGen" }, { "id": "wangp", "name": "WanGP / Wan2GP", "domain": "Images", "kind": "Multi-model runner", "summary": "GPU-oriented application supporting image models alongside video models.", "bestFor": [ "Operating supported open image models in a shared GPU application", "Use when you want a shared interface over open models on suitable hardware." ], "notFor": [ "A model-independent image-quality guarantee", "It does not make Wan, LTX and Hunyuan the same model." ], "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "GPU installation and selected model", "inputs": [ "Prompt", "Model-specific image inputs", "Prompts", "References", "Selected model weights" ], "outputs": [ "Raster images", "Outputs from the chosen engine" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking.", "Memory and output capabilities depend on the model; no blanket laptop-readiness claim." ], "sourceUrls": [ { "title": "WanGP / Wan2GP for images official source", "url": "https://github.com/deepbeepmeep/Wan2GP" }, { "title": "WanGP / Wan2GP official source", "url": "https://github.com/deepbeepmeep/Wan2GP" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Operating supported open image models in a shared GPU application", "stars": 3, "rationale": "Runner can consolidate open-model workflows; model and hardware setup remain separate.", "recommendation": "Consider for this task", "compareTask": "Operate multiple image engines", "domain": "Images" }, { "task": "Video creation workspaces", "stars": 4, "rationale": "A practical orchestration layer; compare engines separately inside it.", "recommendation": "Use when you want a shared interface over open models on suitable hardware.", "domain": "Images", "compareTask": "Video creation workspaces" } ], "relatedRecipePaths": [ "diagramas/index.html", "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/wangp.jpg", "comparisonNotes": "Qwen Image and FLUX are engines it can run, not alternate names for WanGP.", "thumbnailEvidence": "GitHub repository social preview card for deepbeepmeep/Wan2GP, retrieved 14 September 2026.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images", "Video" ], "thumbnailSource": "https://github.com/deepbeepmeep/Wan2GP" }, { "id": "concept-diagrams", "name": "concept-diagrams", "domain": "Images", "kind": "Studio recipe", "summary": "Chooses a conceptual structure and visual material, then generates and judges illustrations.", "bestFor": "An image that explains one concept before words", "notFor": "A factual network or diagram whose connections must be exact", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Use the Studio recipe with a chosen image engine; add real labels separately", "inputs": [ "Concept sentence", "Reference style", "Audience" ], "outputs": [ "Concept illustration", "HTML/SVG labels" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [], "sourceChecked": "2026-09-05", "evidenceLevel": "Local primary source: diagramas/index.html inspected; existing examples, no new test.", "ratings": [ { "task": "An image that explains one concept before words", "stars": 5, "rationale": "Existing Studio examples demonstrate the recipe; it directly matches the house illustration task.", "recommendation": "Use for this task", "compareTask": "Explain a concept through illustration", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/covers/concept-diagrams.svg", "comparisonNotes": "A recipe, not an engine. Its engine can change without renaming the task.", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ] }, { "id": "diagram-design", "name": "diagram-design", "domain": "Images", "kind": "Diagram authoring recipe", "summary": "Writes structured editorial diagrams as self-contained HTML and SVG.", "bestFor": "Static architecture, flows, sequences and timelines", "notFor": "Photorealistic illustration", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Agent skill with factual brief; render and inspect", "inputs": [ "Facts", "Relationships", "Diagram brief" ], "outputs": [ "HTML", "SVG" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "diagram-design official source", "url": "https://github.com/cathrynlavery/diagram-design" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Static architecture, flows, sequences and timelines", "stars": 5, "rationale": "Explicit shapes and labels suit structured technical diagrams.", "recommendation": "Use for this task", "compareTask": "Draw exact systems and processes", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/diagram-design.jpg", "comparisonNotes": "Current repository has expanded beyond the old Studio count. Avoid hardcoding type counts that immediately drift.", "thumbnailEvidence": "Architecture-screenshot thumbnail from the official diagram-design GitHub README, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://github.com/cathrynlavery/diagram-design" }, { "id": "archify", "name": "archify", "domain": "Images", "kind": "Diagram authoring and validation recipe", "summary": "Builds explorable system maps and supports validation and visual exports.", "bestFor": "A system map grounded in repository evidence", "notFor": "An attractive map accepted without checking source facts", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Agent skill and validation/delivery CLI", "inputs": [ "Repository evidence", "System facts", "Relationships" ], "outputs": [ "HTML", "SVG", "PNG", "WebM" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "archify official source", "url": "https://github.com/tt-a1i/archify" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "A system map grounded in repository evidence", "stars": 5, "rationale": "Grounded map and validation workflow are a direct fit for technical explanations.", "recommendation": "Use for this task", "compareTask": "Draw exact systems and processes", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/archify.jpg", "comparisonNotes": "Choose over a decorative image engine when relationships must remain inspectable.", "thumbnailEvidence": "Official hero image from the archify GitHub README, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://github.com/tt-a1i/archify" }, { "id": "mermaid", "name": "Mermaid", "domain": "Images", "kind": "Diagram language and renderer", "summary": "Turns text definitions into diagrams through a JavaScript renderer.", "bestFor": "Maintainable flowcharts and sequence diagrams", "notFor": "Unusual art-directed composition or photo illustration", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Write Mermaid syntax and render in supported host", "inputs": [ "Diagram text" ], "outputs": [ "Rendered diagram", "SVG via compatible exporter" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Mermaid official source", "url": "https://mermaid.js.org/intro/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Maintainable flowcharts and sequence diagrams", "stars": 5, "rationale": "Text-defined relationships are easy for AI to revise and reviewers to inspect.", "recommendation": "Use for this task", "compareTask": "Draw exact systems and processes", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/mermaid.jpg", "comparisonNotes": "Useful default for small structural diagrams; switch to custom SVG for composition requiring precise control.", "thumbnailEvidence": "Official header image (sample rendered diagrams) from the Mermaid GitHub README, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://github.com/mermaid-js/mermaid" }, { "id": "fossflow", "name": "FossFLOW", "domain": "Images", "kind": "Isometric diagram editor and renderer", "summary": "Creates isometric infrastructure diagrams with reusable scene objects.", "bestFor": "Servers, stores and infrastructure shown as an isometric board", "notFor": "A general image-generation engine", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Existing local JSON/render harness or browser editor", "inputs": [ "Scene JSON", "Nodes", "Connections" ], "outputs": [ "Isometric diagram", "PNG through recorded harness" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "FossFLOW official source", "url": "https://github.com/Abrar74774/FossFLOW" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Servers, stores and infrastructure shown as an isometric board", "stars": 5, "rationale": "Its specialised scene vocabulary matches the infrastructure-board task.", "recommendation": "Use for this task", "compareTask": "Draw exact systems and processes", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/fossflow.jpg", "comparisonNotes": "Isometric view is a visual convention, not a reason to use it for every system.", "thumbnailEvidence": "Demo GIF frame from the official FossFLOW GitHub README, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://github.com/Abrar74774/FossFLOW" }, { "id": "figma-images", "name": "Figma Design", "domain": "Images", "kind": "Vector and layout editor", "summary": "Edits design objects and exports selected content as image/vector assets.", "bestFor": "Composing editable assets and complete interface layouts", "notFor": "A substitute for choosing an image generation model", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Figma application and authorised connector actions", "inputs": [ "Shapes", "Text", "Images", "Components" ], "outputs": [ "Editable design", "PNG", "JPG", "SVG", "PDF" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Figma Design official source", "url": "https://help.figma.com/hc/en-us/articles/360040028114-Export-static-designs-from-Figma" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Composing editable assets and complete interface layouts", "stars": 5, "rationale": "Editable layers and export fit production layout; connector capability is distinct from model generation.", "recommendation": "Use for this task", "compareTask": "Compose editable designs", "domain": "Images" } ], "relatedRecipePaths": [ "tools.html" ], "thumbnailCandidate": "catalogue/tool-captures/figma-images.jpg", "comparisonNotes": "Current task exposes Figma-related tooling; old blanket statements that FigJam/Figma always require a human mouse should be revisited.", "thumbnailEvidence": "Official share image from the Figma homepage, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://www.figma.com/" }, { "id": "open-pencil", "name": "OpenPencil", "domain": "Images", "kind": "Programmable design editor", "summary": "Edits design documents through app, CLI and MCP.", "bestFor": "AI-driven editable layout and asset production", "notFor": "Treating an editor as a photo generator", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "CLI or MCP, document operations and export", "inputs": [ "Design file", "Shapes", "Text", "Images" ], "outputs": [ "Editable design", "Exported assets" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "OpenPencil official source", "url": "https://github.com/open-pencil/open-pencil" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "AI-driven editable layout and asset production", "stars": 5, "rationale": "Documented headless inspection/edit/export supports the Studio automation requirement.", "recommendation": "Use for this task", "compareTask": "Compose editable designs", "domain": "Images" } ], "relatedRecipePaths": [ "tools.html" ], "thumbnailCandidate": "catalogue/tool-captures/open-pencil.jpg", "comparisonNotes": "Keep distinct from Pencil.dev; similar names do not imply the same application.", "thumbnailEvidence": "Official application screenshot from the OpenPencil GitHub README, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://github.com/open-pencil/open-pencil" }, { "id": "penpot", "name": "Penpot", "domain": "Images", "kind": "Vector and prototyping editor", "summary": "Design platform with vector editing, layouts and prototyping guidance.", "bestFor": "Collaborative editable design when a Penpot instance is chosen", "notFor": "Adding a second editor without a workflow reason", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Hosted/self-hosted application; validate chosen integration", "inputs": [ "Shapes", "Text", "Images", "Design files" ], "outputs": [ "Editable design", "Exported assets" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Penpot official source", "url": "https://help.penpot.app/user-guide/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Collaborative editable design when a Penpot instance is chosen", "stars": 3, "rationale": "Good editor category fit but explicitly deferred locally.", "recommendation": "Consider for this task", "compareTask": "Compose editable designs", "domain": "Images" } ], "relatedRecipePaths": [ "tools.html" ], "thumbnailCandidate": "catalogue/tool-captures/penpot.jpg", "comparisonNotes": "Respect the recorded defer decision. Do not imply a connector is running just because one exists.", "thumbnailEvidence": "Official share image from the Penpot help-center user guide, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://help.penpot.app/user-guide/" }, { "id": "koboyo", "name": "Koboyo", "domain": "Images", "kind": "Drawing canvas", "summary": "Browser canvas recorded for manual shapes, text and diagram drawing.", "bestFor": "Drawing a quick diagram by hand", "notFor": "Promising an unverified automation interface", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Browser canvas", "inputs": [ "Shapes", "Text", "Images", "Diagram text as recorded" ], "outputs": [ "Canvas diagram" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Koboyo official source", "url": "https://koboyo.com/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Local kit assessment; official landing page reachable but not text-inspectable.", "ratings": [ { "task": "Drawing a quick diagram by hand", "stars": null, "rationale": "Official site opened but returned no readable feature content; only historical local assessment available.", "recommendation": "Not rated; verify before choosing", "compareTask": "Compose editable designs", "domain": "Images" } ], "relatedRecipePaths": [ "tools.html" ], "thumbnailCandidate": "catalogue/tool-captures/koboyo.jpg", "comparisonNotes": "Manual-use suitability is recorded, not retested. No claim that automation is impossible today.", "thumbnailEvidence": "Official share image from the Koboyo homepage, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://koboyo.com/" }, { "id": "unsplash", "name": "Unsplash", "domain": "Images", "kind": "Photography source", "summary": "Source of real photographs under its published licence.", "bestFor": "A real-world photograph for a page", "notFor": "A record of a specific event that the photograph does not document", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Find an individual photo and retain provenance and applicable licence", "inputs": [ "Subject brief", "Selected photo page" ], "outputs": [ "Photograph" ], "limitations": [ "Do not confuse editorial illustration with evidence of an event.", "Check whether the selected asset is covered by the displayed licence and relevant third-party rights." ], "sourceUrls": [ { "title": "Unsplash official source", "url": "https://unsplash.com/license" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "A real-world photograph for a page", "stars": 5, "rationale": "Real photographs directly serve real-world imagery; selection quality still requires human judgment.", "recommendation": "Use for this task", "compareTask": "Find real photographs", "domain": "Images" } ], "relatedRecipePaths": [ "tools.html", "inspiration/index.html" ], "thumbnailCandidate": "catalogue/covers/unsplash.svg", "comparisonNotes": "Source, not model. A representative shelf thumbnail must not be labelled as a tested Unsplash result.", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "costModel": "Check the chosen image and current licence/tier.", "dataLocation": "Photo source; no inference requirement.", "languageSupport": "Not applicable to the image source.", "domains": [ "Images" ] }, { "id": "animos", "name": "Animos", "domain": "Images", "kind": "Image presentation and video exporter", "summary": "Arranges existing pictures into animated display patterns.", "bestFor": "Showing a set of images with a tunnel, globe, spiral or similar pattern", "notFor": "Generating the underlying pictures", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Browser editor; recorded pattern-harvest script for study", "inputs": [ "Existing images" ], "outputs": [ "Animated presentation", "Video export as recorded" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Animos official source", "url": "https://animos.app/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Local 59-pattern shelf inspected previously; official landing opened but provides no readable feature text.", "ratings": [ { "task": "Showing a set of images with a tunnel, globe, spiral or similar pattern", "stars": 4, "rationale": "The Studio already retains pattern examples, making presentation selection tangible.", "recommendation": "Use for this task", "compareTask": "Present a set of images", "domain": "Images" } ], "relatedRecipePaths": [ "inspiration/presenting-images.html" ], "thumbnailCandidate": "catalogue/tool-captures/animos.jpg", "comparisonNotes": "Pattern count and export tier are historical; current pricing and export entitlement are not verified.", "thumbnailEvidence": "Official share image from the Animos homepage, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://animos.app/" }, { "id": "qwen-image", "name": "Qwen-Image family", "domain": "Images", "kind": "Image engine", "summary": "Image generation and editing family with released model code and weights for specific variants.", "bestFor": "Evaluating open image generation and editing", "notFor": "Assuming every advertised variant runs in the installed runner", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Compatible inference stack or supported WanGP integration; choose exact variant", "inputs": [ "Prompt", "Images for edit models" ], "outputs": [ "Raster images" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Qwen-Image family official source", "url": "https://github.com/QwenLM/Qwen-Image" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Evaluating open image generation and editing", "stars": null, "rationale": "Official family capabilities verified; no local evaluation to justify a comparative recommendation.", "recommendation": "Not rated; verify before choosing", "compareTask": "Generate and edit illustrations", "domain": "Images" } ], "relatedRecipePaths": [ "diagramas/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/qwen-image.jpg", "comparisonNotes": "Qwen Image is an engine family, WanGP a runner. Record exact weights, licence and inference implementation.", "thumbnailEvidence": "Official benchmark image from the Qwen-Image GitHub README (hosted on Alibaba's own CDN), retrieved 14 September 2026.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images" ], "thumbnailSource": "https://github.com/QwenLM/Qwen-Image" }, { "id": "meta-image", "name": "Meta Muse Image", "domain": "Images", "kind": "Image engine candidate", "summary": "Image capability mentioned in the existing Studio tool register.", "bestFor": "Watchlist pending primary feature verification", "notFor": "Production selection based on the old register alone", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Not established in this review", "inputs": [ "Not verified" ], "outputs": [ "Not verified" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Meta Muse Image official source", "url": "https://ai.developer.meta.com/docs/overview/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Historical Studio entry only; official overview retrieval did not verify image features.", "ratings": [ { "task": "Watchlist pending primary feature verification", "stars": null, "rationale": "Official page yielded insufficient readable content; capabilities remain unverified.", "recommendation": "Not rated; verify before choosing", "compareTask": "Generate and edit illustrations", "domain": "Images" } ], "relatedRecipePaths": [ "tools.html" ], "thumbnailCandidate": "catalogue/covers/meta-image.svg", "comparisonNotes": "Retain as unrated watchlist rather than silently turning a prior claim into current fact.", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Images" ] }, { "id": "logobook", "name": "Logobook", "domain": "Images", "kind": "Visual reference source", "summary": "Gallery of logos, symbols and trademarks.", "bestFor": "Studying logo structure and visual precedents", "notFor": "Downloading a mark to use as your own", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Browse and record design observations", "inputs": [ "Research question" ], "outputs": [ "References and notes" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Logobook official source", "url": "https://logobook.com/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Studying logo structure and visual precedents", "stars": 4, "rationale": "A focused logo gallery fits visual research, while creation remains a separate step.", "recommendation": "Use for this task", "compareTask": "Study logo references", "domain": "Images" } ], "relatedRecipePaths": [ "inspiration/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/logobook.jpg", "comparisonNotes": "A reference source, not an image generator or a licence to reuse logos.", "thumbnailEvidence": "Headless capture of the official Logobook homepage (its logo directory), retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://logobook.com/" }, { "id": "hugeicons", "name": "Hugeicons", "domain": "Images", "kind": "Icon asset library", "summary": "Library of icons for design and development.", "bestFor": "Selecting a consistent icon vocabulary", "notFor": "Generating a bespoke photograph or treating every icon tier as identically licensed", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Choose matching set; check licence and export/integration", "inputs": [ "Icon concept", "Style choice" ], "outputs": [ "Icon assets" ], "limitations": [ "Suitability is an editorial judgment, not a measured quality or speed ranking." ], "sourceUrls": [ { "title": "Hugeicons official source", "url": "https://hugeicons.com/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official documentation checked; no comparative generation test", "ratings": [ { "task": "Selecting a consistent icon vocabulary", "stars": 4, "rationale": "Existing icon families serve UI consistency better than unrelated generated icons.", "recommendation": "Use for this task", "compareTask": "Choose consistent icons", "domain": "Images" } ], "relatedRecipePaths": [ "inspiration/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/hugeicons.jpg", "comparisonNotes": "Cross-list under UI. Avoid published icon-count totals or plan claims that drift.", "thumbnailEvidence": "Official share image from the Hugeicons homepage, retrieved 14 September 2026.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Images" ], "thumbnailSource": "https://hugeicons.com/" }, { "id": "veo", "name": "Veo", "domain": "Video", "kind": "Hosted video model", "compareTask": "Generate video shots", "summary": "Google video-generation model accessed through supported Google APIs and interfaces.", "bestFor": "Use for a Google-based scripted shot-generation pipeline.", "notFor": "Google AI Studio is an interface; Veo is the model.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompt", "Reference image where supported" ], "outputs": [ "Generated clip" ], "limitations": [ "Pin the exact version and check supported reference/audio features. A matching last frame does not guarantee a seamless edit." ], "sourceUrls": [ { "title": "Veo official source", "url": "https://ai.google.dev/gemini-api/docs/veo" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Generate video shots", "stars": 4, "rationale": "Clear API route for controlled shot generation; no visual-quality lead is claimed.", "recommendation": "Use for a Google-based scripted shot-generation pipeline.", "domain": "Video", "compareTask": "Generate video shots" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/veo.jpg", "thumbnailEvidence": "Crop of the official Google DeepMind Veo product page hero (\"Veo, our leading video generation model\"), retrieved 14 September 2026.", "comparisonNotes": "Google AI Studio is an interface; Veo is the model. Pin the exact version and check supported reference/audio features. A matching last frame does not guarantee a seamless edit.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Video" ], "thumbnailSource": "https://deepmind.google/models/veo/" }, { "id": "seedance", "name": "Seedance", "domain": "Video", "kind": "Hosted video model", "compareTask": "Generate video shots", "summary": "ByteDance video-generation family with multimodal reference capabilities in newer releases.", "bestFor": "Evaluate when reference-driven motion and multi-shot storytelling are important.", "notFor": "Not a complete editor, and versions have different capabilities.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompt", "Image/audio/video references supported by the chosen version" ], "outputs": [ "Generated clip" ], "limitations": [ "Choose the actual model version offered by the provider; no current quota or price claim." ], "sourceUrls": [ { "title": "Seedance official source", "url": "https://seed.bytedance.com/en/blog/one-take-creation-flexible-referencing-introducing-seedance-2-5" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Generate video shots", "stars": 4, "rationale": "A relevant alternative for reference-driven shots; official capability descriptions do not establish superiority on your brief.", "recommendation": "Evaluate when reference-driven motion and multi-shot storytelling are important.", "domain": "Video", "compareTask": "Generate video shots" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/seedance.jpg", "thumbnailEvidence": "Headless capture of the official ByteDance Seed blog post announcing Seedance 2.5, retrieved 14 September 2026.", "comparisonNotes": "Not a complete editor, and versions have different capabilities. Choose the actual model version offered by the provider; no current quota or price claim.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Video" ], "thumbnailSource": "https://seed.bytedance.com/en/blog/one-take-creation-flexible-referencing-introducing-seedance-2-5" }, { "id": "grok-imagine", "name": "Grok Imagine Video", "domain": "Video", "kind": "Hosted video model", "compareTask": "Generate video shots", "summary": "xAI video generation, distinct from Grok’s live conversational voice.", "bestFor": "Use for the existing Grok-based generation route, then inspect identity, speech and editability.", "notFor": "A consumer subscription is not the same entitlement as the API.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompt", "Supported image/video references" ], "outputs": [ "Generated video" ], "limitations": [ "Do not call it free per finished video without accounting for plan limits, retries and the selected access route." ], "sourceUrls": [ { "title": "Grok Imagine Video official source", "url": "https://docs.x.ai/developers/models/grok-imagine-video" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Generate video shots", "stars": 4, "rationale": "Useful because a Studio route already exists; rejected takes and service limits still matter.", "recommendation": "Use for the existing Grok-based generation route, then inspect identity, speech and editability.", "domain": "Video", "compareTask": "Generate video shots" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/grok-imagine.jpg", "thumbnailEvidence": "Official share image from xAI's Grok Imagine Video documentation page, retrieved 14 September 2026.", "comparisonNotes": "A consumer subscription is not the same entitlement as the API. Do not call it free per finished video without accounting for plan limits, retries and the selected access route.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Video" ], "thumbnailSource": "https://docs.x.ai/developers/models/grok-imagine-video" }, { "id": "minimax-h3", "name": "MiniMax H3", "domain": "Video", "kind": "Video/audio model", "compareTask": "Generate video shots", "summary": "MiniMax multimodal video model with reference-based generation.", "bestFor": "Evaluate where mixed reference inputs are needed; select API or deployment route explicitly.", "notFor": "Do not reuse old claims about memory, cost or unrestricted deployment.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Text", "Supported image/video/audio references" ], "outputs": [ "Generated audiovisual clip" ], "limitations": [ "Check the actual model licence and hosting route. Existing geographic/licensing summaries are not legal clearance." ], "sourceUrls": [ { "title": "MiniMax H3 official source", "url": "https://www.minimax.io/blog/minimax-h3" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Generate video shots", "stars": 3, "rationale": "Official model identity confirmed, but the Studio route and results remain untested.", "recommendation": "Evaluate where mixed reference inputs are needed; select API or deployment route explicitly.", "domain": "Video", "compareTask": "Generate video shots" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/minimax-h3.jpg", "thumbnailEvidence": "Official share image from MiniMax's own blog post announcing MiniMax H3, retrieved 14 September 2026.", "comparisonNotes": "Do not reuse old claims about memory, cost or unrestricted deployment. Check the actual model licence and hosting route. Existing geographic/licensing summaries are not legal clearance.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video" ], "thumbnailSource": "https://www.minimax.io/blog/minimax-h3" }, { "id": "hailuo", "name": "Hailuo", "domain": "Video", "kind": "Video-generation application", "compareTask": "Video creation workspaces", "summary": "MiniMax’s user-facing video creation service.", "bestFor": "Choose the app when you want an interactive MiniMax workflow.", "notFor": "Hailuo is the interface; H3 and other models are the engines.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompt", "Reference material" ], "outputs": [ "Generated clips" ], "limitations": [ "Model availability in the app may differ from the API." ], "sourceUrls": [ { "title": "Hailuo official source", "url": "https://github.com/MiniMax-AI/MiniMax-H3/blob/main/README.md" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Video creation workspaces", "stars": 3, "rationale": "Distinguishes the application from the underlying model; account and route need checking.", "recommendation": "Choose the app when you want an interactive MiniMax workflow.", "domain": "Video", "compareTask": "Video creation workspaces" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/hailuo.jpg", "thumbnailEvidence": "Headless capture of the official Hailuo AI (MiniMax) video-creation app homepage, retrieved 14 September 2026.", "comparisonNotes": "Hailuo is the interface; H3 and other models are the engines. Model availability in the app may differ from the API.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ], "thumbnailSource": "https://hailuoai.video/" }, { "id": "kling", "name": "Kling", "domain": "Video", "kind": "Video model/platform family", "compareTask": "Generate video shots", "summary": "Video-generation family named in the existing Studio routes.", "bestFor": "Keep as an alternative for a like-for-like shot test through a verified provider.", "notFor": "LivePortrait is a separate open portrait-animation tool, not the same service.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompt", "Supported references" ], "outputs": [ "Generated clips" ], "limitations": [ "Provider model listing confirms availability only within that provider, not performance or direct-account access." ], "sourceUrls": [], "sourceChecked": null, "evidenceLevel": "Existing Studio reference; current primary-source verification incomplete", "ratings": [ { "task": "Generate video shots", "stars": null, "rationale": "Current direct official documentation could not be read in this pass; no fair task rating assigned.", "recommendation": "Keep as an alternative for a like-for-like shot test through a verified provider.", "domain": "Video", "compareTask": "Generate video shots" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/kling.jpg", "thumbnailEvidence": "Official Kling AI demo image (an image-expansion result) served from Kling's own homepage assets, retrieved 14 September 2026. Shows a real generation result, not a benchmark claim.", "comparisonNotes": "LivePortrait is a separate open portrait-animation tool, not the same service. Provider model listing confirms availability only within that provider, not performance or direct-account access.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video" ], "thumbnailSource": "https://kling.ai/" }, { "id": "wan", "name": "Wan video models", "domain": "Video", "kind": "Open video model family", "compareTask": "Run open video models", "summary": "Open model family with text, image, speech-to-video and animation variants.", "bestFor": "Start here when you need an open-model experiment on a suitable GPU.", "notFor": "WanGP is a runner; Wan is a family of models with different inputs.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompt", "Image or audio depending on variant" ], "outputs": [ "Generated video" ], "limitations": [ "Budget the full workflow including setup, retries and processing; no universal cost-per-minute claim." ], "sourceUrls": [ { "title": "Wan video models official source", "url": "https://github.com/Wan-Video/Wan2.2" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Run open video models", "stars": 4, "rationale": "Multiple explicit model variants make it a useful self-hosted route; choose the task-specific variant.", "recommendation": "Start here when you need an open-model experiment on a suitable GPU.", "domain": "Video", "compareTask": "Run open video models" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/wan.jpg", "thumbnailEvidence": "Computational-efficiency table image from the official Wan2.2 GitHub README, retrieved 14 September 2026.", "comparisonNotes": "WanGP is a runner; Wan is a family of models with different inputs. Budget the full workflow including setup, retries and processing; no universal cost-per-minute claim.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video" ], "thumbnailSource": "https://github.com/Wan-Video/Wan2.2" }, { "id": "ltx", "name": "LTX Video / LTX-2", "domain": "Video", "kind": "Open video model family", "compareTask": "Run open video models", "summary": "Lightricks video models and associated generation workflows.", "bestFor": "Evaluate alongside Wan with the same shot brief and hardware constraints.", "notFor": "LTX Studio, an application, is not identical to the open model repository.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompt", "Supported references" ], "outputs": [ "Generated clips" ], "limitations": [ "Do not transfer runtime or quality claims between LTX versions." ], "sourceUrls": [ { "title": "LTX Video / LTX-2 official source", "url": "https://github.com/Lightricks/LTX-2" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Run open video models", "stars": 4, "rationale": "A distinct open-model alternative; suitability depends on version and integration rather than the family name.", "recommendation": "Evaluate alongside Wan with the same shot brief and hardware constraints.", "domain": "Video", "compareTask": "Run open video models" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/ltx.jpg", "thumbnailEvidence": "GitHub repository social preview card for Lightricks/LTX-2, retrieved 14 September 2026.", "comparisonNotes": "LTX Studio, an application, is not identical to the open model repository. Do not transfer runtime or quality claims between LTX versions.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video" ], "thumbnailSource": "https://github.com/Lightricks/LTX-2" }, { "id": "hunyuan-video", "name": "HunyuanVideo", "domain": "Video", "kind": "Open video model family", "compareTask": "Run open video models", "summary": "Tencent open video model family.", "bestFor": "Use as a candidate where its selected model and deployment meet the project needs.", "notFor": "Not a ready local application simply because weights are published.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompt", "References supported by the chosen model" ], "outputs": [ "Generated clips" ], "limitations": [ "Check model-specific licence, hardware and runtime support." ], "sourceUrls": [ { "title": "HunyuanVideo official source", "url": "https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Run open video models", "stars": 3, "rationale": "Relevant for experimentation; no working Studio deployment established here.", "recommendation": "Use as a candidate where its selected model and deployment meet the project needs.", "domain": "Video", "compareTask": "Run open video models" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/hunyuan-video.jpg", "thumbnailEvidence": "GitHub repository social preview card for Tencent-Hunyuan/HunyuanVideo-1.5, retrieved 14 September 2026.", "comparisonNotes": "Not a ready local application simply because weights are published. Check model-specific licence, hardware and runtime support.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video" ], "thumbnailSource": "https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5" }, { "id": "google-ai-studio", "name": "Google AI Studio", "domain": "Video", "kind": "Model-development interface", "compareTask": "Video creation workspaces", "summary": "Google interface for exploring and building with its models.", "bestFor": "Use to try the supported Google model and prepare an API workflow.", "notFor": "Do not rate it against Veo as if it were another video model.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Prompts", "Supported media" ], "outputs": [ "Model responses or generated media" ], "limitations": [ "Check the available model and account tier rather than assuming access from old tutorials." ], "sourceUrls": [ { "title": "Google AI Studio official source", "url": "https://ai.google.dev/gemini-api/docs/video" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Video creation workspaces", "stars": 4, "rationale": "The natural interface counterpart to a Google-model route; output quality comes from the selected model.", "recommendation": "Use to try the supported Google model and prepare an API workflow.", "domain": "Video", "compareTask": "Video creation workspaces" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/google-ai-studio.jpg", "thumbnailEvidence": "Official share image from the Google AI Studio product page on ai.google.dev, retrieved 14 September 2026.", "comparisonNotes": "Do not rate it against Veo as if it were another video model. Check the available model and account tier rather than assuming access from old tutorials.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video" ], "thumbnailSource": "https://ai.google.dev/aistudio" }, { "id": "heygen", "name": "HeyGen", "domain": "Video", "kind": "Hosted avatar platform", "compareTask": "Create a speaking presenter", "summary": "Hosted API platform for creating avatar videos and related presenter workflows.", "bestFor": "Choose managed avatar production when you want a hosted script-to-presenter route.", "notFor": "Not an open local lip-sync model.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Presenter/avatar", "Text or audio" ], "outputs": [ "Speaking presenter video" ], "limitations": [ "Different avatar, live-avatar, translation and lip-sync endpoints have different requirements." ], "sourceUrls": [ { "title": "HeyGen official source", "url": "https://docs.heygen.com/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official product/developer source checked; no comparative output test", "ratings": [ { "task": "Create a speaking presenter", "stars": 4, "rationale": "Official developer overview verified. No new presenter-quality test was run.", "recommendation": "Choose managed avatar production when you want a hosted script-to-presenter route.", "domain": "Video", "compareTask": "Create a speaking presenter" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/heygen.jpg", "thumbnailEvidence": "Official share image from the HeyGen homepage, retrieved 14 September 2026.", "comparisonNotes": "Different avatar, live-avatar, translation and lip-sync endpoints have different requirements.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Video", "Speakers" ], "thumbnailSource": "https://www.heygen.com/" }, { "id": "liveportrait", "name": "LivePortrait", "domain": "Video", "kind": "Portrait-animation model", "compareTask": "Animate an existing portrait", "summary": "Transfers driving motion to a source portrait with retargeting controls.", "bestFor": "Use when the desired motion already exists in a driving source.", "notFor": "It is not itself a narration generator or a complete audio-driven speaking pipeline.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Portrait", "Driving image/video motion" ], "outputs": [ "Animated portrait" ], "limitations": [ "Do not promise a spoken performance from a photograph and text alone." ], "sourceUrls": [ { "title": "LivePortrait official source", "url": "https://github.com/KlingAIResearch/LivePortrait" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Animate an existing portrait", "stars": 4, "rationale": "Strong task fit for motion transfer; extra speech/lip-sync steps are separate.", "recommendation": "Use when the desired motion already exists in a driving source.", "domain": "Video", "compareTask": "Animate an existing portrait" } ], "relatedRecipePaths": [ "people/factory.html", "people/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/liveportrait.jpg", "thumbnailEvidence": "Showcase GIF frame from the official LivePortrait GitHub README, retrieved 14 September 2026.", "comparisonNotes": "It is not itself a narration generator or a complete audio-driven speaking pipeline. Do not promise a spoken performance from a photograph and text alone.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video", "Speakers" ], "thumbnailSource": "https://github.com/KlingAIResearch/LivePortrait" }, { "id": "musetalk", "name": "MuseTalk", "domain": "Video", "kind": "Lip-sync model", "compareTask": "Lip-sync an existing face clip", "summary": "Audio-driven lip synchronisation applied to a face image/video route.", "bestFor": "Use when you already have the face material and approved speech to synchronise.", "notFor": "Not a full-body motion generator or a replacement for voice synthesis.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Face image/video", "Approved audio" ], "outputs": [ "Lip-synchronised video" ], "limitations": [ "Real-time claims depend on hardware and setup; no measured speed claim here." ], "sourceUrls": [ { "title": "MuseTalk official source", "url": "https://github.com/TMElyralab/MuseTalk" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Lip-sync an existing face clip", "stars": 4, "rationale": "Directly matches the lip-sync stage; inspect mouth detail and identity on your material.", "recommendation": "Use when you already have the face material and approved speech to synchronise.", "domain": "Video", "compareTask": "Lip-sync an existing face clip" } ], "relatedRecipePaths": [ "people/factory.html", "people/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/musetalk.jpg", "thumbnailEvidence": "Gradio demo screenshot from the official MuseTalk GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Not a full-body motion generator or a replacement for voice synthesis. Real-time claims depend on hardware and setup; no measured speed claim here.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video", "Speakers" ], "thumbnailSource": "https://github.com/TMElyralab/MuseTalk" }, { "id": "latentsync", "name": "LatentSync", "domain": "Video", "kind": "Lip-sync diffusion model", "compareTask": "Lip-sync an existing face clip", "summary": "Audio-conditioned lip synchronisation for existing video.", "bestFor": "Compare against MuseTalk using the same face clip and approved audio.", "notFor": "Not a complete talking-person workflow by itself.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Face video", "Approved audio" ], "outputs": [ "Lip-synchronised video" ], "limitations": [ "Version changes affect memory and output; more guidance can distort the face." ], "sourceUrls": [ { "title": "LatentSync official source", "url": "https://github.com/bytedance/LatentSync" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Lip-sync an existing face clip", "stars": 4, "rationale": "A true same-stage alternative to MuseTalk; final choice needs visual inspection, not unrelated model scores.", "recommendation": "Compare against MuseTalk using the same face clip and approved audio.", "domain": "Video", "compareTask": "Lip-sync an existing face clip" } ], "relatedRecipePaths": [ "people/factory.html", "people/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/latentsync.jpg", "thumbnailEvidence": "Model architecture diagram from the official ByteDance LatentSync GitHub README, retrieved 14 September 2026.", "comparisonNotes": "Not a complete talking-person workflow by itself. Version changes affect memory and output; more guidance can distort the face.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video", "Speakers" ], "thumbnailSource": "https://github.com/bytedance/LatentSync" }, { "id": "ffmpeg", "name": "FFmpeg", "domain": "Video", "kind": "Media processing tool", "compareTask": "Assemble, convert and export", "summary": "Command-line processing for video, audio, filters, subtitles and export.", "bestFor": "Default for repeatable assembly, transcoding and technical finishing.", "notFor": "It does not invent footage or replace editorial judgement.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Command-line tool", "inputs": [ "Existing media", "Timing and filter instructions" ], "outputs": [ "Edited/converted media" ], "limitations": [ "Check output resolution, audio, subtitle timing and codecs on the finished file." ], "sourceUrls": [ { "title": "FFmpeg official source", "url": "https://ffmpeg.org/ffmpeg.html" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Assemble, convert and export", "stars": 5, "rationale": "Deterministic and scriptable for the established finishing layer; complements generative tools.", "recommendation": "Default for repeatable assembly, transcoding and technical finishing.", "domain": "Video", "compareTask": "Assemble, convert and export" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/ffmpeg.jpg", "thumbnailEvidence": "Headless capture of the top of the official FFmpeg homepage, retrieved 14 September 2026.", "comparisonNotes": "It does not invent footage or replace editorial judgement. Check output resolution, audio, subtitle timing and codecs on the finished file.", "costModel": "Local compute; no metered generation service.", "dataLocation": "Local file processing unless remote inputs/outputs are configured.", "languageSupport": "Processes media and subtitle streams; does not translate speech by itself.", "domains": [ "Video" ], "thumbnailSource": "https://ffmpeg.org/" }, { "id": "remotion", "name": "Remotion", "domain": "Video", "kind": "Code-based video renderer", "compareTask": "Assemble, convert and export", "summary": "Creates videos from React code: text, charts and layouts placed frame by frame, rendered to MP4.", "bestFor": "Choose for repeatable explainers with precise text, charts and branded composition.", "notFor": "Not a generative video model or a conventional timeline editor.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Code and renderer. With Claude: add Remotion's official skills (npx skills add remotion-dev/skills), then ask Claude for the video; it creates the project, previews it and renders the MP4.", "inputs": [ "Code", "Images", "Audio", "Data" ], "outputs": [ "Rendered video" ], "limitations": [ "Requires code, Node.js and a headless Chrome that Remotion downloads on first render.", "On a Mac, macOS 15 or later.", "Free under the Remotion License for individuals, non-profit organisations and companies of up to three employees; larger for-profit companies need a company licence." ], "sourceUrls": [ { "title": "Remotion documentation", "url": "https://www.remotion.dev/docs/" }, { "title": "Remotion source code", "url": "https://github.com/remotion-dev/remotion" }, { "title": "Remotion agent skills", "url": "https://www.remotion.dev/docs/ai/skills" }, { "title": "Remotion License", "url": "https://github.com/remotion-dev/remotion/blob/main/LICENSE.md" } ], "sourceChecked": "2026-09-14", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Assemble, convert and export", "stars": 5, "rationale": "Best fit in this set for controlled programmatic layouts and reusable explainer templates.", "recommendation": "Choose for repeatable explainers with precise text, charts and branded composition.", "domain": "Video", "compareTask": "Assemble, convert and export" } ], "relatedRecipePaths": [ "film.html", "family-video.html", "product-films.html" ], "thumbnailCandidate": "catalogue/tool-captures/remotion.png", "thumbnailEvidence": "Frame from a three-second test render made with Remotion's blank starter on 14 September 2026. Shows the kind of output, not a quality comparison.", "comparisonNotes": "Not a generative video model or a conventional timeline editor. Requires code and a rendering environment. Free under the Remotion License for individuals, non-profit organisations and companies of up to three employees; larger for-profit companies need a company licence.", "costModel": "Own compute. Free under the Remotion License for individuals, non-profit organisations and companies of up to three employees; larger for-profit companies need a company licence.", "dataLocation": "Rendering deployment determines where media is processed.", "languageSupport": "Renders supplied content; language depends on text/fonts and any external voice engine.", "domains": [ "Video" ], "thumbnailSource": "https://github.com/remotion-dev/remotion" }, { "id": "final-cut-pro", "name": "Final Cut Pro", "domain": "Video", "kind": "Timeline editor", "compareTask": "Edit a film by hand", "summary": "Apple video-editing application for assembling and finishing a film.", "bestFor": "Use when a person needs precise editorial control over timing, picture and sound.", "notFor": "Not interchangeable with a code renderer or a footage generator.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Human-operated editing application", "inputs": [ "Video", "Audio", "Titles" ], "outputs": [ "Edited film" ], "limitations": [ "Current licence and installation not checked." ], "sourceUrls": [ { "title": "Final Cut Pro official source", "url": "https://www.apple.com/final-cut-pro/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Edit a film by hand", "stars": 4, "rationale": "A strong manual finishing option; choose it for editor control rather than automated batch generation.", "recommendation": "Use when a person needs precise editorial control over timing, picture and sound.", "domain": "Video", "compareTask": "Edit a film by hand" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/final-cut-pro.jpg", "thumbnailEvidence": "Official share image from Apple's Final Cut Pro product page, retrieved 14 September 2026.", "comparisonNotes": "Not interchangeable with a code renderer or a footage generator. Current licence and installation not checked.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ], "thumbnailSource": "https://www.apple.com/final-cut-pro/" }, { "id": "openshot", "name": "OpenShot", "domain": "Video", "kind": "Timeline editor", "compareTask": "Edit a film by hand", "summary": "Open-source video editor with timeline editing and effects.", "bestFor": "Choose for accessible manual editing without adopting a commercial editor.", "notFor": "Not a high-level prompt-to-film engine.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Human-operated editing application", "inputs": [ "Video", "Audio", "Images" ], "outputs": [ "Edited video" ], "limitations": [ "Check export and performance with the intended project." ], "sourceUrls": [ { "title": "OpenShot official source", "url": "https://www.openshot.org/features/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Edit a film by hand", "stars": 3, "rationale": "A useful lighter manual option; select by needed editing features and workflow.", "recommendation": "Choose for accessible manual editing without adopting a commercial editor.", "domain": "Video", "compareTask": "Edit a film by hand" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/openshot.jpg", "thumbnailEvidence": "Official share image from the OpenShot features page, retrieved 14 September 2026.", "comparisonNotes": "Not a high-level prompt-to-film engine. Check export and performance with the intended project.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ], "thumbnailSource": "https://www.openshot.org/features/" }, { "id": "capcut", "name": "CapCut", "domain": "Video", "kind": "Timeline/content editor", "compareTask": "Edit a film by hand", "summary": "Desktop video editor with editing, effects, search and AI-assisted content features.", "bestFor": "Choose for quick manual editing with effects and social-format delivery.", "notFor": "Not a substitute for a controlled scripted render.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Video", "Audio", "Text" ], "outputs": [ "Edited video" ], "limitations": [ "Check feature availability, plan restrictions and export settings in the current application." ], "sourceUrls": [ { "title": "CapCut official source", "url": "https://www.capcut.com/tools/desktop-video-editor" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official product/developer source checked; no comparative output test", "ratings": [ { "task": "Edit a film by hand", "stars": 4, "rationale": "Strong fit for interactive editing; not a model-quality comparison.", "recommendation": "Choose for quick manual editing with effects and social-format delivery.", "domain": "Video", "compareTask": "Edit a film by hand" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/capcut.jpg", "thumbnailEvidence": "Official share image from the CapCut desktop video editor page, retrieved 14 September 2026.", "comparisonNotes": "Check feature availability, plan restrictions and export settings in the current application.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ], "thumbnailSource": "https://www.capcut.com/tools/desktop-video-editor" }, { "id": "real-esrgan", "name": "Real-ESRGAN", "domain": "Video", "kind": "Restoration model", "compareTask": "Restore and upscale media", "summary": "Image/video restoration and upscaling project.", "bestFor": "Use when existing footage needs restoration, then inspect whether details remain faithful.", "notFor": "It cannot reliably recover facts absent from the source.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Existing frames/images" ], "outputs": [ "Restored/upscaled frames" ], "limitations": [ "Check faces, fine text and temporal consistency; restoration can invent detail." ], "sourceUrls": [ { "title": "Real-ESRGAN official source", "url": "https://github.com/xinntao/Real-ESRGAN" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Restore and upscale media", "stars": 4, "rationale": "A specialised finishing step, separate from generation and editing.", "recommendation": "Use when existing footage needs restoration, then inspect whether details remain faithful.", "domain": "Video", "compareTask": "Restore and upscale media" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/real-esrgan.jpg", "thumbnailEvidence": "Teaser comparison image from the official Real-ESRGAN GitHub README, retrieved 14 September 2026.", "comparisonNotes": "It cannot reliably recover facts absent from the source. Check faces, fine text and temporal consistency; restoration can invent detail.", "costModel": "Compute and model/weight terms; hosted wrappers charge separately.", "dataLocation": "Self-hosted if deployed locally, otherwise provider-dependent.", "languageSupport": "Exact model version determines support.", "domains": [ "Video" ], "thumbnailSource": "https://github.com/xinntao/Real-ESRGAN" }, { "id": "moneyprinterturbo", "name": "MoneyPrinterTurbo", "domain": "Video", "kind": "Automated video workflow", "compareTask": "Automate a content pipeline", "summary": "Builds short videos through a pipeline of scripts, assets, narration and assembly.", "bestFor": "Use for prototyping a repeatable short-video pipeline with reviewed inputs.", "notFor": "It is orchestration, not a new video-generation model.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Topic or script", "Providers and source assets" ], "outputs": [ "Assembled short video" ], "limitations": [ "Provider dependencies and asset rights require explicit selection; do not treat affiliate provider claims as project evidence." ], "sourceUrls": [ { "title": "MoneyPrinterTurbo official source", "url": "https://github.com/harry0703/MoneyPrinterTurbo" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Automate a content pipeline", "stars": 3, "rationale": "Useful starting automation; factual accuracy, media relevance and editorial quality still need review.", "recommendation": "Use for prototyping a repeatable short-video pipeline with reviewed inputs.", "domain": "Video", "compareTask": "Automate a content pipeline" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/moneyprinterturbo.jpg", "thumbnailEvidence": "WebUI screenshot from the official MoneyPrinterTurbo GitHub README, retrieved 14 September 2026.", "comparisonNotes": "It is orchestration, not a new video-generation model. Provider dependencies and asset rights require explicit selection; do not treat affiliate provider claims as project evidence.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ], "thumbnailSource": "https://github.com/harry0703/MoneyPrinterTurbo" }, { "id": "runpod", "name": "RunPod", "domain": "Video", "kind": "GPU hosting service", "compareTask": "Host a video model", "summary": "Infrastructure for running GPU workloads.", "bestFor": "Use when an open model needs hardware beyond the local computer.", "notFor": "Not a model and not a video-quality competitor.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Container/runtime", "Model", "Inputs" ], "outputs": [ "Compute execution and model outputs" ], "limitations": [ "Include startup, idle time, storage, failed jobs and retries in any cost calculation." ], "sourceUrls": [ { "title": "RunPod official source", "url": "https://docs.runpod.io/overview" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Host a video model", "stars": 4, "rationale": "Fits the existing rented-GPU architecture; compare full workflow cost rather than advertised GPU rate alone.", "recommendation": "Use when an open model needs hardware beyond the local computer.", "domain": "Video", "compareTask": "Host a video model" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/runpod.jpg", "thumbnailEvidence": "Official share image from the RunPod homepage, retrieved 14 September 2026.", "comparisonNotes": "Not a model and not a video-quality competitor. Include startup, idle time, storage, failed jobs and retries in any cost calculation.", "costModel": "Rented compute, storage and runtime; include retries and idle time.", "dataLocation": "Cloud GPU hosting.", "languageSupport": "Determined by the model you deploy.", "domains": [ "Video" ], "thumbnailSource": "https://www.runpod.io/" }, { "id": "yt-dlp", "name": "yt-dlp", "domain": "Video", "kind": "Media acquisition tool", "compareTask": "Acquire source media", "summary": "Command-line downloader for supported media sources.", "bestFor": "Use to obtain permitted source media for analysis or editing.", "notFor": "It does not create footage, edit a film or bypass access rights.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Supported media URL" ], "outputs": [ "Local media file" ], "limitations": [ "Supported sites change. Confirm the right to use acquired media." ], "sourceUrls": [ { "title": "yt-dlp official source", "url": "https://github.com/yt-dlp/yt-dlp" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official capability documentation; no comparative production benchmark", "ratings": [ { "task": "Acquire source media", "stars": 5, "rationale": "The established acquisition layer; complements FFmpeg for processing.", "recommendation": "Use to obtain permitted source media for analysis or editing.", "domain": "Video", "compareTask": "Acquire source media" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/yt-dlp.jpg", "thumbnailEvidence": "GitHub repository social preview card for yt-dlp/yt-dlp, retrieved 14 September 2026.", "comparisonNotes": "It does not create footage, edit a film or bypass access rights. Supported sites change. Confirm the right to use acquired media.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ], "thumbnailSource": "https://github.com/yt-dlp/yt-dlp" }, { "id": "magic-hour", "name": "Magic Hour", "domain": "Video", "kind": "Hosted media platform", "compareTask": "Video creation workspaces", "summary": "Hosted platform grouping text/image/video generation and editing workflows.", "bestFor": "Choose a focused hosted task such as photo-to-video or restyling when it fits your source material.", "notFor": "Do not confuse a reference, workflow or vendor with a generation engine.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Task-specific source material" ], "outputs": [ "Depends on the identified workflow" ], "limitations": [ "The platform routes different underlying models; compare the actual model and access route." ], "sourceUrls": [ { "title": "Magic Hour official source", "url": "https://magichour.ai/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official product/developer source checked; no comparative output test", "ratings": [ { "task": "Video creation workspaces", "stars": 3, "rationale": "A documented task-oriented workspace; no current Studio production trial establishes a default.", "recommendation": "Choose a focused hosted task such as photo-to-video or restyling when it fits your source material.", "domain": "Video", "compareTask": "Video creation workspaces" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/magic-hour.jpg", "thumbnailEvidence": "Official share image from the Magic Hour homepage, retrieved 14 September 2026.", "comparisonNotes": "The platform routes different underlying models; compare the actual model and access route.", "costModel": "Hosted plan or usage charges; current price not quoted.", "dataLocation": "Provider-hosted processing; check the selected route.", "languageSupport": "Coverage varies by model/feature; test the required language.", "domains": [ "Video" ], "thumbnailSource": "https://magichour.ai/" }, { "id": "palmier", "name": "Palmier Pro", "domain": "Video", "kind": "Video workflow application", "compareTask": "Automate a content pipeline", "summary": "Mac timeline editor with generation and agent editing through MCP.", "bestFor": "Evaluate for human-and-agent timeline editing if the machine meets its requirements.", "notFor": "Do not confuse a reference, workflow or vendor with a generation engine.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Task-specific source material" ], "outputs": [ "Depends on the identified workflow" ], "limitations": [ "Official requirements specify macOS 26 on Apple Silicon. Editing/MCP and paid generation have different access conditions." ], "sourceUrls": [ { "title": "Palmier Pro official source", "url": "https://github.com/palmier-io/palmier-pro" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official product/developer source checked; no comparative output test", "ratings": [ { "task": "Automate a content pipeline", "stars": 3, "rationale": "Distinctive agent-editable timeline; conditional rating because current local compatibility is not verified.", "recommendation": "Evaluate for human-and-agent timeline editing if the machine meets its requirements.", "domain": "Video", "compareTask": "Automate a content pipeline" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/palmier.png", "thumbnailEvidence": "Project screenshot from the official source, retrieved 5 September 2026. Not a comparative quality test.", "comparisonNotes": "Official requirements specify macOS 26 on Apple Silicon. Editing/MCP and paid generation have different access conditions.", "costModel": "Editor and MCP separate from paid generative features.", "dataLocation": "Local timeline; AI generation uses external providers.", "languageSupport": "Transcription and generation depend on the selected engines.", "domains": [ "Video" ], "thumbnailSource": "https://raw.githubusercontent.com/palmier-io/palmier-pro/main/assets/palmier-ui.png" }, { "id": "wondershare", "name": "Wondershare (product unresolved)", "domain": "Video", "kind": "Vendor label", "compareTask": "Unresolved identities", "summary": "Vendor name in the existing material; the exact editor/product is not specified.", "bestFor": "Retain as a reference; resolve current capability and access before selection.", "notFor": "Do not confuse a reference, workflow or vendor with a generation engine.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Task-specific source material" ], "outputs": [ "Depends on the identified workflow" ], "limitations": [ "Current primary-source verification or exact identity remains incomplete." ], "sourceUrls": [], "sourceChecked": null, "evidenceLevel": "Existing Studio reference; current primary-source verification incomplete", "ratings": [ { "task": "Unresolved identities", "stars": null, "rationale": "Coverage retained without inventing a rating; evidence is not sufficient for a current recommendation.", "recommendation": "Retain as a reference; resolve current capability and access before selection.", "domain": "Video", "compareTask": "Unresolved identities" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/covers/wondershare.svg", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "comparisonNotes": "Do not confuse a reference, workflow or vendor with a generation engine. Current primary-source verification or exact identity remains incomplete.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ] }, { "id": "video-toolkit", "name": "Studio video toolkit", "domain": "Video", "kind": "Internal orchestration recipe", "compareTask": "Automate a content pipeline", "summary": "The local collection of scripts and skills that connects production steps.", "bestFor": "Retain as a reference; resolve current capability and access before selection.", "notFor": "Do not confuse a reference, workflow or vendor with a generation engine.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Task-specific source material" ], "outputs": [ "Depends on the identified workflow" ], "limitations": [ "Current primary-source verification or exact identity remains incomplete." ], "sourceUrls": [], "sourceChecked": null, "evidenceLevel": "Existing Studio reference; current primary-source verification incomplete", "ratings": [ { "task": "Automate a content pipeline", "stars": null, "rationale": "Coverage retained without inventing a rating; evidence is not sufficient for a current recommendation.", "recommendation": "Retain as a reference; resolve current capability and access before selection.", "domain": "Video", "compareTask": "Automate a content pipeline" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/covers/video-toolkit.svg", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "comparisonNotes": "Do not confuse a reference, workflow or vendor with a generation engine. Current primary-source verification or exact identity remains incomplete.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ] }, { "id": "stillslab", "name": "StillsLab", "domain": "Video", "kind": "Visual reference library", "compareTask": "Find a visual direction", "summary": "Film and television frame library for selecting composition, lighting and visual references.", "bestFor": "Use to choose and describe a visual direction before generating a shot.", "notFor": "Do not confuse a reference, workflow or vendor with a generation engine.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Task-specific source material" ], "outputs": [ "Depends on the identified workflow" ], "limitations": [ "Reference frames remain third-party works; do not treat them as production stock assets." ], "sourceUrls": [ { "title": "StillsLab official source", "url": "https://stillslab.com/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official product/developer source checked; no comparative output test", "ratings": [ { "task": "Find a visual direction", "stars": 4, "rationale": "Useful reference source, not a generator; there is no generated-image quality score.", "recommendation": "Use to choose and describe a visual direction before generating a shot.", "domain": "Video", "compareTask": "Find a visual direction" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/stillslab.jpg", "thumbnailEvidence": "Official share image from the StillsLab homepage, retrieved 14 September 2026.", "comparisonNotes": "Reference frames remain third-party works; do not treat them as production stock assets.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ], "thumbnailSource": "https://stillslab.com/" }, { "id": "motionsites", "name": "MotionSites", "domain": "Video", "kind": "Reference/prompt library", "compareTask": "Find a visual direction", "summary": "Website design and motion-prompt reference library.", "bestFor": "Use for motion-led website references and prompts; keep film production in its own workflow.", "notFor": "Do not confuse a reference, workflow or vendor with a generation engine.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Application/API or supported runtime; access not rechecked", "inputs": [ "Task-specific source material" ], "outputs": [ "Depends on the identified workflow" ], "limitations": [ "A prompt or reference is not a working editor or an assurance of a reproducible result." ], "sourceUrls": [ { "title": "MotionSites official source", "url": "https://motionsites.ai/" } ], "sourceChecked": "2026-09-05", "evidenceLevel": "Official product/developer source checked; no comparative output test", "ratings": [ { "task": "Find a visual direction", "stars": 4, "rationale": "Strong fit as a website-motion reference, not a general-purpose film tool.", "recommendation": "Use for motion-led website references and prompts; keep film production in its own workflow.", "domain": "Video", "compareTask": "Find a visual direction" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/motionsites.jpg", "thumbnailEvidence": "Official share image from the MotionSites homepage, retrieved 14 September 2026.", "comparisonNotes": "A prompt or reference is not a working editor or an assurance of a reproducible result.", "costModel": "Check application, model and compute charges separately.", "dataLocation": "Depends on the selected application, backend and hosting route.", "languageSupport": "Check the selected model or feature; no blanket language guarantee.", "domains": [ "Video" ], "thumbnailSource": "https://motionsites.ai/" }, { "id": "hand-drawn-explainer-nikola", "name": "Hand-drawn explainer (Nikola)", "domain": "Video", "kind": "Agent skill for explainer videos", "compareTask": "Automate a content pipeline", "summary": "A Codex skill that turns a topic, a script or a subtitle file into an explainer video drawn stroke by stroke, or animated as cards and diagrams.", "bestFor": "Choose when an idea should be drawn in front of the viewer, one part at a time, with subtitles and an editable project.", "notFor": "Not a quick template tool: it is written for Chinese narration and calls outside voice and image services.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Codex agent skill with Python scripts; its animated route renders with HyperFrames", "inputs": [ "Topic", "Script", "SRT subtitles", "Source picture" ], "outputs": [ "MP4 video", "Subtitles", "Timeline", "Editable project" ], "limitations": [ "Documentation and default narration are Chinese; the default voice is ByteDance Volcengine speech synthesis and needs its account.", "The stroke route draws from a source picture that has to be made first, for example with an image model." ], "sourceUrls": [ { "title": "Repository and examples", "url": "https://github.com/hi-nikola/hand-drawn-explainer-video-nikola" } ], "sourceChecked": "2026-09-13", "evidenceLevel": "Repository documentation and published example videos; no run on this Mac", "ratings": [ { "task": "Automate a content pipeline", "stars": 3, "rationale": "The only explainer in this set drawn stroke by stroke on one sheet, delivered as a real MP4 with subtitles; held back by Chinese-first narration and outside services.", "recommendation": "Choose when an idea should be drawn in front of the viewer, one part at a time.", "domain": "Video", "compareTask": "Automate a content pipeline" } ], "relatedRecipePaths": [ "film.html", "diagramas/index.html" ], "thumbnailCandidate": "catalogue/covers/hand-drawn-explainer-nikola.svg", "thumbnailSource": "https://raw.githubusercontent.com/hi-nikola/hand-drawn-explainer-video-nikola/HEAD/examples/stroke-story/yuefa-sanzhang/seven-scenes-final-frames.jpg", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "comparisonNotes": "Drawn stroke by stroke on one sheet rather than cut between shots. Its animated route uses HyperFrames, which renders HTML to MP4.", "costModel": "Free code under Apache-2.0; the voice and image services it calls are billed by their providers.", "dataLocation": "Scripts run locally; narration and image generation go to the selected outside services.", "languageSupport": "Written for Chinese narration; another language needs another voice service.", "domains": [ "Video" ] }, { "id": "video-shotcraft", "name": "Video Shotcraft", "domain": "Video", "kind": "Agent skill for product films", "compareTask": "Automate a content pipeline", "summary": "A Claude Code and Codex skill that turns a product's real screens into a short promo film on Remotion, using 157 shot recipes with tuned code, sound effects and optional music.", "bestFor": "Choose for a product or service film built from its own screens, when exact text and an editable result matter more than a speaker.", "notFor": "Not for narration, speakers or generated footage: it adds no voice.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Claude Code or Codex agent skill: it captures the product's screens with Puppeteer, codes each shot in Remotion, reviews the cut and opens a browser editor.", "inputs": [ "Product URL or local server", "Screenshots", "Copy and brand styles", "Music (optional)" ], "outputs": [ "MP4 with and without music", "Remotion project", "Browser-editable timeline", "JianYing draft (optional)" ], "limitations": [ "Skill instructions are in Chinese.", "No voice or speaker; narration is a separate project.", "Defaults to a fast, beat-cut marketing register.", "Its delivery steps open a browser and ask to credit the author and submit the film to its showcase.", "Several bundled sound effects have no traced licence.", "Run nothing posted in its issue tracker: issue 66 hides a command that sends a machine's environment variables to an outside server." ], "sourceUrls": [ { "title": "Repository", "url": "https://github.com/Vincentwei1021/video-shotcraft" }, { "title": "Online gallery", "url": "https://vincentwei1021.github.io/video-shotcraft/" } ], "sourceChecked": "2026-09-14", "evidenceLevel": "Repository code, CI checks and the published gallery; no run on this Mac", "ratings": [ { "task": "Automate a content pipeline", "stars": 4, "rationale": "The only pipeline in this set that builds the film from the product's own screens, on Remotion, with no paid service; held back by Chinese instructions and no voice.", "recommendation": "Choose for a screen-based product film without a speaker.", "domain": "Video", "compareTask": "Automate a content pipeline" } ], "relatedRecipePaths": [ "film.html" ], "thumbnailCandidate": "catalogue/tool-captures/video-shotcraft.jpg", "thumbnailSource": "https://vincentwei1021.github.io/video-shotcraft/", "thumbnailEvidence": "Capture of the project's online gallery, 14 September 2026.", "comparisonNotes": "Builds on Remotion and adds the shot library, a 2.5D camera over real screenshots, 149 sound effects, a review step and a browser editor.", "costModel": "Free code under Apache-2.0; rendering runs on Remotion, free for individuals, non-profit organisations and companies of up to three employees.", "dataLocation": "Capture, coding and rendering run locally.", "languageSupport": "Skill instructions in Chinese; on-screen text comes from the product and the copy supplied.", "domains": [ "Video" ] }, { "id": "infinitetalk", "name": "InfiniteTalk", "domain": "Video", "kind": "Open audio-driven avatar model", "compareTask": "Create a speaking presenter", "summary": "Turns a portrait or a clip and an approved audio track into a speaking video with head and body motion. Open weights from MeiGen-AI.", "bestFor": "The cheapest measured route for a portrait speaking our own audio: identity held, about $2.30 a published minute at 480p on WaveSpeed.", "notFor": "Not for sealed lips on b, p and m in French from a small portrait: four variants failed Frank's eye on 14-09-2026.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Hosted API (WaveSpeed $0.03 a second at 480p) or open weights in WanGP on a rented card; measured 14-09-2026", "inputs": [ "Portrait image or clip", "Approved audio", "Optional prompt" ], "outputs": [ "Speaking video, 480p or 720p" ], "limitations": [ "The lips do not seal on the b of « de bord », even with the real voice, at 720p or with the word slowed (14-09-2026).", "The last word of a clip is under-articulated; a silent tail of about 1 s, cut afterwards, helps.", "Self-hosted, the 14B model ran out of memory on a 46.5 GB RunPod container on 20-08-2026." ], "sourceUrls": [ { "title": "InfiniteTalk repository", "url": "https://github.com/MeiGen-AI/InfiniteTalk" }, { "title": "WaveSpeed API and price", "url": "https://wavespeed.ai/docs/docs-api/wavespeed-ai/infinitetalk" } ], "sourceChecked": "2026-09-15", "evidenceLevel": "Measured on our own speaker 14-09-2026: six hosted clips, cost and wall clock read, judged by Frank", "ratings": [ { "task": "Create a speaking presenter", "stars": 4, "rationale": "The only born-whole route measured on our own speaker: identity held, about 70 s and $0.09 to $0.12 for a 3 to 4 s clip at 480p. Lip shapes on bilabials failed Frank's eye.", "recommendation": "Start here for a moving speaker on approved audio; check the b, p and m frames before approving.", "domain": "Video", "compareTask": "Create a speaking presenter" } ], "relatedRecipePaths": [ "people/factory.html", "people/index.html" ], "comparisonNotes": "Same family as LongCat-Video-Avatar: the whole person is generated from portrait and audio, not a mouth repainted on an existing clip.", "costModel": "WaveSpeed $0.03 a second at 480p, $0.06 at 720p (read 14-09-2026); measured $0.12 for 3.1 s.", "dataLocation": "Provider-hosted on WaveSpeed; self-hosted on a rented card.", "languageSupport": "French tested by us: identity and rhythm pass, bilabial closures fail at 480p.", "domains": [ "Video", "Speakers" ], "thumbnailCandidate": "catalogue/tool-captures/infinitetalk.jpg", "thumbnailEvidence": "Frame from an official demo video in the InfiniteTalk GitHub README (a model output, its watermark kept), retrieved 15 September 2026.", "thumbnailSource": "https://github.com/MeiGen-AI/InfiniteTalk" }, { "id": "longcat-video-avatar", "name": "LongCat-Video-Avatar 1.5", "domain": "Video", "kind": "Open audio-driven avatar model", "compareTask": "Create a speaking presenter", "summary": "Turns one portrait and an approved audio track into a speaking video, face and body born together. Open weights from Meituan.", "bestFor": "A portrait speaking our own approved audio, when InfiniteTalk misses lip shapes.", "notFor": "Not runnable on the Mac (NVIDIA cards only), and not tested by its authors in French or Spanish.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Hosted API (WaveSpeed $0.04 a second at 480p, fal $0.15) or open weights on a rented card; not run by us yet", "inputs": [ "Portrait image", "Approved audio", "Optional prompt" ], "outputs": [ "Speaking video, 480p or 720p, up to 2 minutes a job on WaveSpeed" ], "limitations": [ "The model card evaluates English and Chinese only.", "Self-hosting needs torch 2.6 and, in the authors' examples, two cards; the 20-08-2026 bench never got it built.", "WaveSpeed warns that jobs longer than 1 minute may be unstable." ], "sourceUrls": [ { "title": "LongCat-Video repository", "url": "https://github.com/meituan-longcat/LongCat-Video" }, { "title": "LongCat-Video-Avatar-1.5 model card", "url": "https://huggingface.co/meituan-longcat/LongCat-Video-Avatar-1.5" }, { "title": "WaveSpeed API and price", "url": "https://wavespeed.ai/docs/docs-api/wavespeed-ai/longcat-avatar" }, { "title": "fal API and price", "url": "https://fal.ai/models/fal-ai/longcat-single-avatar/image-audio-to-video" } ], "sourceChecked": "2026-09-15", "evidenceLevel": "Official model card and provider prices read 15-09-2026; no output test on our speakers", "ratings": [ { "task": "Create a speaking presenter", "stars": 3, "rationale": "The strongest open claims on lip sync (Whisper-Large audio encoder, 8-step distillation, MIT weights), but not yet run on our material. The rating moves after the French bilabial test on « tableau de bord ».", "recommendation": "Test it on the lip shapes InfiniteTalk misses before using it for a whole speaker.", "domain": "Video", "compareTask": "Create a speaking presenter" } ], "relatedRecipePaths": [ "people/factory.html", "people/index.html" ], "comparisonNotes": "Same family as InfiniteTalk: the whole person is generated from portrait and audio, not a mouth repainted on an existing clip.", "costModel": "WaveSpeed $0.04 a second at 480p, $0.08 at 720p, 3-second minimum; fal $0.15 and $0.30 (read 15-09-2026). Open weights: rented card time.", "dataLocation": "Provider-hosted on WaveSpeed or fal; self-hosted on a rented card.", "languageSupport": "Authors evaluated English and Chinese; French untested.", "domains": [ "Video", "Speakers" ], "thumbnailCandidate": "catalogue/tool-captures/longcat-video-avatar.jpg", "thumbnailEvidence": "Frame from the official demo clip assets/2.mp4 on the LongCat-Video-Avatar-1.5 model card (a model output), retrieved 15 September 2026.", "thumbnailSource": "https://huggingface.co/meituan-longcat/LongCat-Video-Avatar-1.5" }, { "id": "video-use", "name": "video-use", "domain": "Video", "kind": "Claude Code editing skill", "compareTask": "Cut filmed speech by its words", "summary": "A Claude Code skill that edits talking footage from its transcript: picks takes, cuts filler words and dead air, grades, subtitles and renders with ffmpeg.", "bestFor": "Raw takes of people speaking (a recorded speaker, a demo, an interview) cut into one clean film by asking Claude.", "notFor": "Not for montages built on music (use family-video), and not for private footage as shipped: the audio goes to ElevenLabs.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Claude Code skill on the Mac; local transcription tested 15-09-2026, ElevenLabs Scribe optional at $0.22 an hour of audio", "inputs": [ "Folder of raw video takes", "A plain-English brief" ], "outputs": [ "edit/final.mp4 with optional subtitles, colour grade and overlays", "A cut list (edl.json) and a project.md memory" ], "limitations": [ "With Scribe, every clip's audio is uploaded to ElevenLabs; the local helper keeps it on the Mac but hears one speaker only and misses some fillers and brand names.", "Burned subtitles need ffmpeg with libass; the Homebrew build has none.", "Cuts follow speech; footage without speech gives it little to work with." ], "sourceUrls": [ { "title": "video-use repository", "url": "https://github.com/browser-use/video-use" }, { "title": "ElevenLabs API prices", "url": "https://elevenlabs.io/pricing/api" } ], "sourceChecked": "2026-09-15", "evidenceLevel": "Run 15-09-2026 on a 33 s French recording: local transcript, packed phrases and a two-phrase cut rendered; subtitle burn-in not run (ffmpeg without libass)", "ratings": [ { "task": "Cut filmed speech by its words", "stars": 4, "rationale": "The only tool here that cuts at word boundaries, with production checks built into its render (30 ms fades at each cut, subtitles applied last, every cut checked in the rendered file). Tested locally on French speech; subtitles not yet burned.", "recommendation": "Use for talking footage; swap Scribe for the local faster-whisper before any private clip.", "domain": "Video", "compareTask": "Cut filmed speech by its words" } ], "relatedRecipePaths": [ "film.html" ], "comparisonNotes": "Beside family-video, which arranges photos and clips on music, and Video Shotcraft, which films a product's own screens: video-use cuts what people say.", "costModel": "ElevenLabs Scribe $0.22 an hour of audio (read 15-09-2026); rendering runs locally.", "dataLocation": "Local by default on this Mac (faster-whisper); audio goes to ElevenLabs only when Scribe is chosen.", "languageSupport": "Scribe covers more than 90 languages, French and Spanish included.", "domains": [ "Video" ], "thumbnailCandidate": "catalogue/tool-captures/video-use.jpg", "thumbnailEvidence": "Official banner from the video-use GitHub repository: a Claude Code session running the skill's task list, retrieved 15 September 2026.", "thumbnailSource": "https://github.com/browser-use/video-use" }, { "id": "flapkit", "name": "Flapkit", "domain": "UI", "kind": "React component library", "compareTask": "Put a moving display on a page", "summary": "Split-flap departure board for React: letters and digits flip over to their new value, with sound.", "bestFor": "Choose inside a React product when a short value must be seen changing, the way an airport board changes.", "notFor": "Not a chart, a table or a general animation library, not for long text, and not for a house single-file page.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "npm package flapkit, imported into a React 19 page. The structural stylesheet and one look are imported explicitly; nothing is injected at runtime.", "inputs": [ "Short strings, one character per cell", "A character deck, built in or your own", "A motion adapter: cascade or riffle, CSS or Canvas" ], "outputs": [ "An animated split-flap board inside the page", "Mechanical flip sound, optional" ], "limitations": [ "React and React DOM 19 or later are required, the package is ESM, and its tooling needs Node 22.18 or later.", "The board needs JavaScript to appear, so it must never carry a fact the reader depends on.", "Every row repeats the first row's cell structure; another layout needs another board.", "The Canvas renderer restricts arbitrary CSS on the cells; use the CSS engine when the face needs them.", "Version 0.1.0, first published 31 August 2026. No Studio page or product uses it yet." ], "sourceUrls": [ { "title": "Flapkit repository and README", "url": "https://github.com/thecuvii/flapkit" }, { "title": "Flapkit documentation and live demo", "url": "https://cuvii.dev/flapkit/" } ], "sourceChecked": "2026-09-17", "evidenceLevel": "README and official documentation site read, live demo opened and captured on 17 September 2026. Not yet built into a page here.", "ratings": [ { "task": "Put a moving display on a page", "stars": 3, "rationale": "The documented fit for its one job is strong, and the licence and browser test matrix are public. It is conditional here because it only runs inside a React 19 product; a house single-file page cannot use it, and nothing here has been built with it yet.", "recommendation": "Use it in a React product where one figure or status should be seen changing. For a studio page, keep to CSS.", "domain": "UI", "compareTask": "Put a moving display on a page" } ], "relatedRecipePaths": [ "motion/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/flapkit.jpg", "thumbnailEvidence": "Official documentation demo at cuvii.dev/flapkit, captured 17 September 2026.", "comparisonNotes": "The only interface building block in the catalogue so far; it is compared against nothing yet.", "costModel": "Free and MIT licensed. No account, no service, no per-use charge.", "dataLocation": "Runs entirely in the reader's browser.", "languageSupport": "Any characters placed in a deck. The built-in decks are Latin letters, digits and punctuation; the demo also flips Chinese characters.", "domains": [ "UI" ], "thumbnailSource": "https://cuvii.dev/flapkit/" }, { "id": "cube-motion", "name": "Cube", "domain": "UI", "kind": "JavaScript animation library", "compareTask": "Make elements enter, leave and change smoothly", "summary": "Four ready motions for page elements: rise (entrance), leave (exit), morph (a label or icon changing state) and reveal (rise when scrolled into view). Durations, curves and staggers are fixed by the library.", "bestFor": "Choose when a page needs calm entrances, exits and state changes without tuning durations and curves by hand, including on a house single-file page.", "notFor": "Not for scroll-scrubbed scenes, timelines, 3D or anything that needs its own timing; not for text a reader must see before the script loads.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "One import from the npm package cube-motion, or from jsDelivr on a plain HTML page (6.4 KB module). Adapters for React, Vue, Solid and Svelte. Built on the browser's Web Animations API, no dependencies.", "inputs": [ "The elements to move, by selector or node", "Optional: animate the children one after another" ], "outputs": [ "Entrance, exit, state-change and on-scroll motion inside the page" ], "limitations": [ "reveal keeps a block invisible until it is scrolled into view, so a full-page capture, a PDF print or a reader without JavaScript can meet an empty section. Use rise and morph freely; put reveal only on blocks that carry no fact the reader depends on.", "Version 0.1.0, first published on npm on 18 September 2026, one author. Pin the version when loading it from the CDN.", "Loading from jsDelivr is an outside request; a studio page should keep a local copy of the module (MIT allows it).", "The four motions cannot be retimed. That is the point of the library, and also its limit." ], "sourceUrls": [ { "title": "Cube site, API and live demo", "url": "https://cube-motion.dev/" }, { "title": "Cube repository (MIT)", "url": "https://github.com/danielwh2/cube-motion" } ], "sourceChecked": "2026-09-21", "evidenceLevel": "Measured 21 September 2026 in headless Chromium on a plain HTML page loading cube-motion 0.1.0 from jsDelivr. rise ran three staggered animations at 640 ms with 70 ms between them on the curve cubic-bezier(0.2, 0, 0, 1), as documented. With reduced motion requested, the 12 px lift was dropped and only the fade remained. reveal held the list at opacity 0 until it entered the viewport, then showed it. leave and morph were not exercised.", "ratings": [ { "task": "Make elements enter, leave and change smoothly", "stars": 4, "rationale": "It runs on a house single-file page with one import and no build step, its reduced-motion handling is real, and its fixed timings match the calm motion the house rules ask for. One star held back: a week old, one author, and reveal hides content from captures and print.", "recommendation": "Use rise for a page's first view and morph for a button or status that changes. Use reveal sparingly and never on a block that carries a fact.", "domain": "UI", "compareTask": "Make elements enter, leave and change smoothly" } ], "relatedRecipePaths": [ "motion/index.html" ], "thumbnailCandidate": "catalogue/tool-captures/cube-motion.jpg", "thumbnailEvidence": "Official site at cube-motion.dev, captured 21 September 2026.", "comparisonNotes": "Beside Flapkit on the interface shelf. Flapkit is one display for React 19 products; Cube is general page motion that also works on a plain HTML page.", "costModel": "Free and MIT licensed. No account, no service, no per-use charge.", "dataLocation": "Runs entirely in the reader's browser.", "languageSupport": "Language-independent. morph keeps shared leading letters when a label changes, so it suits any script written left to right; right-to-left labels were not tested.", "domains": [ "UI" ], "watch": { "title": "Watch the launch film on X", "url": "https://x.com/dwhitedesign/status/2102014506192851303", "thumbnail": "catalogue/tool-captures/cube-motion-film.jpg", "caption": "35 seconds, silent, by Daniel White, the library's author, who drew every frame as SVG in JavaScript. The picture is one frame of it; the film plays on X." }, "thumbnailSource": "https://cube-motion.dev/" }, { "id": "map3d", "name": "map3d", "domain": "Images", "kind": "3D map generator (web app, open source)", "compareTask": "Make a 3D map of a real place", "summary": "Turns any area of the world into a 3D city map: buildings extruded to their recorded height and roads, from OpenStreetMap, exportable as a GLB 3D model.", "bestFor": "Choose for a 3D view of a real city, district, port or corridor that a page or a film can show, turn or fly over.", "notFor": "Not for exact building heights or surveys, not for textured or photographic cities, and not for a flat map with labels (use a map library or a drawn diagram).", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Open the web app, choose the area, then view it or export a GLB file. The code is React Three Fiber (three.js) and can be run or embedded in a React project.", "inputs": [ "An area on the world map", "OpenStreetMap building and road data, fetched by the app" ], "outputs": [ "A 3D city map in the browser", "A GLB 3D model file for Blender, three.js, a page or a film" ], "limitations": [ "Built from OpenStreetMap, so some building heights are missing or wrong; the README says it cannot guarantee accuracy.", "Plain volumes only: no building textures, materials or terrain height yet (all still on its roadmap).", "Last updated in May 2026; one main author.", "Not tried here: no area exported, no GLB opened." ], "sourceUrls": [ { "title": "map3d web app", "url": "https://map.fleet.im/" }, { "title": "map3d repository and README (MIT)", "url": "https://github.com/cartesiancs/map3d" } ], "sourceChecked": "2026-09-21", "evidenceLevel": "README and repository read 21 September 2026; the web app answered. No area exported and no GLB opened here.", "ratings": [ { "task": "Make a 3D map of a real place", "stars": 3, "rationale": "The only tool here for a 3D model of a real place, free and MIT licensed, with a GLB export that Remotion, three.js or Blender can use. Three stars until one export is tried, because heights come from OpenStreetMap and are sometimes missing.", "recommendation": "Use it for a 3D establishing view of a city, port or border corridor. Check the heights of the buildings that matter before showing it.", "domain": "Images", "compareTask": "Make a 3D map of a real place" } ], "relatedRecipePaths": [], "thumbnailCandidate": "catalogue/tool-captures/map3d.jpg", "thumbnailEvidence": "Project screenshot from the official README, retrieved 21 September 2026.", "comparisonNotes": "Keywords: 3D map, city map, digital twin, buildings, roads, OpenStreetMap, OSM, GLB, three.js, React Three Fiber, map of a city, port, corridor, satellite view in 3D.", "costModel": "Free and MIT licensed. No per-use charge.", "dataLocation": "Runs in the browser; map data comes from OpenStreetMap servers.", "languageSupport": "Not relevant: the output is a 3D model without text.", "domains": [ "Images" ], "thumbnailSource": "https://github.com/cartesiancs/map3d" }, { "relatedRecipePaths": [ "product-films.html" ], "thumbnailCandidate": "catalogue/covers/hyperframes.svg", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "sourceChecked": "2026-09-27", "domains": [ "Video" ], "id": "hyperframes", "name": "HyperFrames", "domain": "Video", "kind": "Code-based video renderer", "compareTask": "Assemble, convert and export", "summary": "HeyGen's open-source framework that turns an HTML page with timed animations into an MP4. Claude writes each scene as a web page and renders it; a change edits one line and re-renders.", "bestFor": "Motion-graphics explainers and product films when the scenes are closer to web pages than to video, especially reusing a site's own HTML and CSS.", "notFor": "Not a generative video model and not a timeline editor. Not needed for the family video, which already runs on Remotion.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Command-line tool plus a browser studio; renders with headless Chrome and FFmpeg on your own machine.", "inputs": [ "HTML, CSS and JavaScript scenes", "Images", "Video", "Audio" ], "outputs": [ "Rendered MP4 video" ], "limitations": [ "Needs Node.js, headless Chrome and FFmpeg.", "Studio has not rendered a test with it yet.", "Released March 2026 and changing fast: pin a version for a project." ], "sourceUrls": [ { "title": "HyperFrames source code", "url": "https://github.com/heygen-com/hyperframes" }, { "title": "HyperFrames site", "url": "https://hyperframes.heygen.com/" } ], "evidenceLevel": "Official repository and site reviewed; no Studio test render", "ratings": [ { "task": "Assemble, convert and export", "stars": 4, "rationale": "Same job as Remotion with a plainer format: HTML instead of React, a friendlier Apache-2.0 licence and no company-size limit. Below Remotion only because Remotion is proven in delivered Studio work and HyperFrames is not.", "recommendation": "Try it on one motion-graphics film built from an existing web page. Keep Remotion for the family video.", "domain": "Video", "compareTask": "Assemble, convert and export" } ], "comparisonNotes": "Rival to Remotion. Same promise: every frame is exact and a change is a re-render, not a restart. Licence is simpler than Remotion's, which is free only up to three employees for companies.", "costModel": "Free, own computer.", "dataLocation": "Local render; nothing leaves the machine.", "languageSupport": "Renders any text you give it." }, { "relatedRecipePaths": [ "product-films.html" ], "thumbnailCandidate": "catalogue/covers/21st-dev.svg", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "sourceChecked": "2026-09-27", "domains": [ "UI" ], "id": "21st-dev", "name": "21st", "domain": "UI", "kind": "Interface component library", "compareTask": "Use ready-made interface components", "summary": "A library of over 10,000 React and Tailwind interface components (buttons, cards, menus, sections) made by design engineers, searchable from Claude through its own connector.", "bestFor": "Showing a believable product screen inside a motion-graphics film, or a React page that needs polished parts instead of improvised ones.", "notFor": "Not for Studio pages or eR pages that follow the house design system, and not useful without React and Tailwind.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Website, command line and a Claude Code plugin that adds the 21st connector.", "inputs": [ "A description of the component you need" ], "outputs": [ "React and Tailwind component code" ], "limitations": [ "React and Tailwind only.", "Generating new components beyond the free allowance is paid.", "Community components vary in quality: pick, don't trust." ], "sourceUrls": [ { "title": "21st site", "url": "https://21st.dev/" }, { "title": "21st MCP for coding agents", "url": "https://21st.dev/mcp" }, { "title": "21st MCP source code", "url": "https://github.com/21st-dev/magic-mcp" } ], "evidenceLevel": "Official site and repository reviewed; not used in Studio work", "ratings": [ { "task": "Use ready-made interface components", "stars": 3, "rationale": "Large, well-made catalogue that stops Claude inventing fake-looking buttons, but tied to React and Tailwind, while Studio and eR pages use their own design system.", "recommendation": "Browse the free site for a film's product screens; add the connector only if React pages become regular work.", "domain": "UI", "compareTask": "Use ready-made interface components" } ], "comparisonNotes": "Solves a real problem in AI-made product films: improvised interface parts look cheap. Remocn plays a similar role for Remotion animations.", "costModel": "Free to browse and copy; connector generations about 5 free a month, Pro about 20 USD a month.", "dataLocation": "Hosted catalogue; the code you copy runs wherever you put it.", "languageSupport": "Interface text is whatever you write." }, { "relatedRecipePaths": [ "product-films.html" ], "thumbnailCandidate": "catalogue/covers/whatships.svg", "thumbnailEvidence": "Capability map: documented role and input → output. Not a screenshot or a measured model result.", "sourceChecked": "2026-09-27", "domains": [ "Video" ], "id": "whatships", "name": "What Ships", "domain": "Video", "kind": "Reference video catalogue", "compareTask": "Find a visual direction", "summary": "A free, curated catalogue of about 2,200 startup launch and demo videos posted on X, sorted by category, with duration and poster frames.", "bestFor": "Picking one or two motion-graphics films whose pace, type and transitions Claude should copy before it builds a video.", "notFor": "Not a download source: it links to the original posts. Launch videos, not family or documentary styles.", "availability": "Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.", "operation": "Browse on the web and hand Claude the link of the film to match.", "inputs": [ "A category or product type" ], "outputs": [ "Links to reference videos" ], "limitations": [ "Watch only, the videos stay on X.", "Style range is startup marketing." ], "sourceUrls": [ { "title": "What Ships", "url": "https://whatships.com/" } ], "evidenceLevel": "Site reviewed 27 September 2026", "ratings": [ { "task": "Find a visual direction", "stars": 3, "rationale": "Naming a reference film is what moves Claude off its default look (centred text, gradient, everything fading in). Good for product films; narrow for anything else.", "recommendation": "Use it for step one of the motion-graphics card: choose a reference, then ask Claude to match it.", "domain": "Video", "compareTask": "Find a visual direction" } ], "comparisonNotes": "A reference is worth more than a description: it gives the pace, the type and the transitions.", "costModel": "Free.", "dataLocation": "Public website.", "languageSupport": "Mostly English videos." } ] }