kyutai-labs/moshi — Moshi is a speech-text foundation model and full-duplex spoken…

A GitHub repository starred by Agentik OS and classified by AGK Intelligence as Models & inference.

kyutai-labs/moshi — 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.

By kyutai-labs on GitHub · starred 148 days ago · 11.4k GitHub stars · 1.1k forks · view the repository on GitHub. Repository page: moshi.

More Models & inference repositories

• p-e-w/heretic — Training-free language model modification: Remove censorship, suppress…
• Edge0-AI/Edge0 — Run 35B MoE models in ~2.5 GB of RAM. Weights stream from SSD; a…
• NVlabs/Sana — SANA & Sol: Efficient Generative Models and Full-Stack Training &…
• microsoft/BitNet — Official inference framework for 1-bit LLMs
• Tiiny-AI/PowerInfer — High-speed Large Language Model Serving for Local Deployment
• xinyao27/jevonian — One local endpoint. The right model for every turn — enforced in…
• muellerberndt/cadence — A brain that learns from live experience. System 1 by default:…
• BerriAI/litellm — The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+…

Every repository Agentik OS stars on GitHub is read and classified by AGK Intelligence: what it is, which market it belongs to, how useful and how mature. 314 repositories, 10 markets. The stars are read every 5 minutes — last update 2026-11-06 17:50 UTC. Full method.