From: kyutai-labs <github.com/kyutai-labs>Newsgroups: alt.freeweights.toolsSubject:

Moshi

Release: rustymimi-0.2.2 (2024-09-22)Commit: c7439c1License: Apache-2.0

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

speechrustymimi-0.2.2releaseMoshiPythonApache-2.0

FactsLicenseApache-2.0LanguagePythonKindspeechRelease2024-09-22PlatformsLinux (GPU), macOS (Apple silicon)Links to the projectpermissive

$ pip install -U moshi

Overview Install Use with Freeweights License Provenance

Overview

Real-time spoken dialogue model with PyTorch and MLX runtimes.

Kind: speech
License: Apache-2.0
Language: Python
Platforms: Linux (GPU), macOS (Apple silicon)
Latest release: rustymimi-0.2.2 (2024-09-22)
Repository: kyutai-labs/moshi

Freeweights links to the project and does not host tool binaries.

Install

$ pip install -U moshi

$ pip install -U moshi

Checked against the project's README on 2026-11-15. The project's own instructions win when they differ: install docs

Freeweights links to the project and does not host tool binaries.

Use with Freeweights

Run a catalog model

Download the weights from Freeweights first (HF_ENDPOINT or fw.py), then point the tool at the folder.

Catalog models it can run (315)

Audio8-TTS-Preview-0.6b Audio8-TTS-Realistic-Air BUD-E-Whisper BUD-E-Whisper_V1.1 chatterbox-4bit chatterbox-8bit chatterbox-fp16 chatterbox-GGUF chatterbox-multilingual-v3 chatterbox-ONNX chatterbox-turbo chatterbox-turbo-4bit chatterbox-turbo-8bit chatterbox-turbo-fp16 chatterbox-turbo-GGUF CosyVoice2-0.5B csm-1b csm-1b-8bit Dia-1.6B Dia-1.6B distil-large-v2 distil-large-v3 distil-large-v3.5 distil-large-v3.5-ct2 distil-large-v3.5-ONNX distil-medium.en distil-medium.en_timestamped distil-small.en distil-small.en distil-small.en_timestamped distil-whisper-large-v3_openvino_int8 Fun-CosyVoice3-0.5B-2512 granite-4.0-1b-speech-ONNX granite-speech-3.2-8b granite-speech-3.3-2b granite-speech-3.3-8b granite-speech-4.1-2b-plus granite-speech-5.0-2b granite-speech-5.0-2b-ONNX hubert-xlarge-ls960-ft kb-whisper-base kb-whisper-base-ONNX kb-whisper-large kb-whisper-large-ONNX kb-whisper-medium kb-whisper-medium-ONNX kb-whisper-small kb-whisper-small-ONNX kb-whisper-tiny kb-whisper-tiny-ONNX kitten-tts-micro-0.8 kitten-tts-mini-0.8 kitten-tts-nano-0.1 kitten-tts-nano-0.1-ONNX kitten-tts-nano-0.8-fp32 kotoba-whisper-bilingual-v1.0 kotoba-whisper-bilingual-v1.0-ONNX kotoba-whisper-v2.2 kotoba-whisper-v2.2-ONNX MiMo-V2.5-ASR

and 255 more.

License

License: Apache-2.0
Class: Permissive

Read the license

Freeweights links to the project and does not host tool binaries.

Provenance

From: kyutai-labs <github.com/kyutai-labs>Newsgroups: alt.freeweights.toolsSubject: MoshiRelease: rustymimi-0.2.2 (2024-09-22)Commit: c7439c1License: Apache-2.0

Repository: kyutai-labs/moshi
Pinned release: rustymimi-0.2.2, 2024-09-22
Release commit: c7439c13e4c68a376aea976227c9694354692ab8
Default branch: main e6a55d2722a65870ef52a6c9f6ecfc0e90f38362
Fetched: 2026-11-15
Takedowns: None.

Freeweights links to the project and does not host tool binaries.

Get it

Install Project repository

Works with

315 catalog models. See the list