Source
Moshi official documentation
Read the primary source.
Reference illustration, not a screenshot of this source.
Moshi
Dialogue model and framework · Voice
Speech-text model and full-duplex spoken-dialogue research framework.
Use it for
Evaluate when research-level control over full-duplex dialogue is the objective.
Choose something else for
Not the simplest production route for a website guide.
Give it
Live audio
Get back
Spoken dialogue
How it is operated
Self-hosted model/runtime
Availability
Provider capabilities and historical Studio notes are distinguished below. Current account access, credits and runtime readiness were not tested.
How charging works
Compute and model/weight terms; hosted wrappers charge separately.
Where processing happens
Self-hosted if deployed locally, otherwise provider-dependent.
Language support
Exact model version determines support.
Open dialogue models
★★★☆☆ 3/5 for this task
Distinct from a modular ASR-LLM-TTS stack; useful to evaluate, with substantial deployment work.
Decision: Evaluate when research-level control over full-duplex dialogue is the objective.
Limits and things to check
- Validate languages, hardware and behaviour on the intended conversation.
Not the simplest production route for a website guide. Validate languages, hardware and behaviour on the intended conversation.
Evidence
Official capability documentation; no listening benchmark
Catalogue review: 2026-09-05. The evidence note identifies what was verified. This date does not imply a fresh tool test or verification of every linked source.
