Moshi 7B

Overview

Kyutai's full-duplex speech model — 7.6B parameters with sub-second latency (~200ms) and two voices, Moshiko and Moshika. A speech architecture, not a text LLM.

When to pick this model

• You're building real-time voice interfaces and need full-duplex behavior
• You need low-latency speech-to-speech without separate TTS and STT
• You're researching speech architectures rather than text LLMs
• You can run inference directly in PyTorch or via Kyutai's stack

VRAM requirements by quantization

Quantization	VRAM required
Q4_K_M (recommended)	5 GB
Q5_K_M	6 GB
Q8_0	9 GB
FP16 (no quantization)	15 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

Strengths

• First open full-duplex speech model
• Sub-second latency (~200ms in practice)
• Mimi codec at 12.5 Hz / 1.1 kbps on 24 kHz audio
• From Kyutai, a respected French AI lab

Limitations

• Not a text LLM — different use case entirely
• Architecture not supported by Ollama
• CC-BY 4.0 license — attribution required

Architecture & training

Architecture: Full-duplex speech-text · Depth Transformer (codebook) + 7B Temporal Transformer

Training: Mimi codec at 12.5 Hz / 1.1 kbps on 24 kHz audio. ~200ms practical latency.

Quick start

# GitHub : kyutai-labs/moshi — voix Moshiko (H) / Moshika (F)

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.