Qwen 3 Omni 30B-A3B

Overview

Alibaba's omni-modal 30B MoE (3B active) with streaming speech, 119-language ASR, and Apache 2.0 licensing. The most accessible truly omnimodal open model.

When to pick this model

• Voice-first assistants with low-latency speech in/out
• Multilingual ASR across 119 languages
• Real-time multimodal agents on a single GPU
• Long-context multimodal reasoning (131k)
• Apache 2.0 commercial deployments

VRAM requirements by quantization

Quantization	VRAM required
Q4_K_M (recommended)	19 GB
Q5_K_M	23 GB
Q8_0	35 GB
FP16 (no quantization)	62 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

• Native omnimodal I/O: text, image, audio in and out
• 131k context
• Streaming speech for low-latency voice apps
• Apache 2.0 license
• Only 3B active params per token

Limitations

• Around 19 GB VRAM in Q4
• Audio path is still maturing relative to text and vision
• Tooling support uneven outside vLLM

Architecture & training

Architecture: MoE · 30B · Qwen3-Omni · text + vision + audio end-to-end

Training: Qwen3-Omni 30B — Qwen omnimodal model (text, images, audio in/out).

Quick start

ollama run qwen3-omni:30b

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