Model reference · open weights

Qwen3-Omni

Qwen3-Omni is an open-weight language model from marksverdhei, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.

Licence fee required LLMs marksverdhei 1 variants 51k downloads/mo
Request a licence + hosting quote All served models Not on the shared API today — deployed on request.

About

What Qwen3-Omni is

Qwen3-Omni-30B-A3B-FP8 Block-wise FP8 quantization of Qwen/Qwen3-Omni-30B-A3B-Instruct. Model Details Components Quantized to FP8: - Thinker (48 layers MoE) - main language model - Talker (20 layers MoE) - audio generation model Kept in BF16: - Vision encoder (thinker.visual) - Audio tower (thinker.audiotower) - Code2Wav decoder (code2wav) - Embedding layers - LayerNorm layers - MoE gate routing layers Usage with vLLM Requirements - vLLM = 0.13.0 with Qwen3-Omni support - 2x 24GB GPUs (e.g., RTX 3090) or equivalent - ~35 GB disk space Quantization Details Block-wise quantization with 128x128 blocks provides better precision than per-tensor quantization while maintaining good compression. Each block has its own scale factor stored as weightscaleinv (inverse scale for efficient multiplication during inference). Original Model This is a quantized version of Qwen/Qwen3-Omni-30B-A3B-Instruct. Qwen3-Omni is a natively end-to-end multilingual omni-modal foundation model that processes text, images, audio, and video, and delivers real-time streaming responses in both text and natural speech. Key Features: - State-of-the-art across modalities - Supports 119 text languages, 19 speech input languages, and 10 speech output languages - MoE-based Thinker-Talker architecture - Real-time audio/video interaction For full details, see the original model card.

Summarised from the published model card. Read the full card on the HuggingFace links below.

Specifications

What it is

Makermarksverdhei
TypeLanguage models
Parameters (lead)35.3B
Variants1
Runs withtransformers
Based onQwen/Qwen3-Omni-30B-A3B-Instruct
Released2026-01-20
Popularity51k downloads / month
Likes4
LicenceCommercial licence needed

How it works

How language models work

Your prompttext / messagesTransformerattention over tokensNext-token loopgenerate + streamResponsetext · tool callsA language model reads your tokens and predicts the next one, again and again, streaming the reply back.

Variants

Sizes & precisions

Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.

VariantParamsPrecisionVRAMFits 16 GBWeights
Qwen3-Omni-30B-A3B-FP835.3BFP8~40.5 GBWeights ↗

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys marksverdhei-qwen3-omni for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (marksverdhei-qwen3-omni below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/chat/completions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"marksverdhei-qwen3-omni","messages":[{"role":"user","content":"Hello"}]}'

Details

Languages, data & research

Languages

en zh ko ja de ru it fr es pt ms nl id tr

Tags

transformers safetensors qwen3_omni_moe text-to-audio multimodal fp8 quantized qwen3 omni any-to-any en zh ko ja

Licence

Commercial licence needed

The weights are open but its licence needs a commercial agreement for business use. AxForge can arrange that licence and host the model for you — you pay AxForge, we settle with the model’s maker. Ask us for a quote. Read the licence ↗

Sources

Weights & code

Want Qwen3-Omni on EU-owned hardware?

Request a licence + hosting quote See what’s served now

Explore

More language models

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms