Qwen/qwen3-8-omni-flash/Released Sep 2026

Qwen3.8 Omni Flash

Omnimodal audio and video understanding, 1M context

Commercial use
Text

About Qwen3.8 Omni Flash

Qwen3.8 Omni Flash is a native omnimodal model that reads text, images, audio, and video in one request and answers in text. It is built for long-form audio-visual understanding, meeting and call analysis, subtitling, and agentic work over recorded media, and it accepts 113 languages and dialects as audio input.

The distinctive behavior is agentic long-form understanding: rather than processing a whole recording, the model starts from the question and gathers evidence in coarse-to-fine passes, so most of the media is never read. Thinking is enabled by default with adjustable reasoning effort, and the model supports function calling and web search. Audio input can be multichannel, letting it use spatial information to separate speakers.

Output is text only, so the model cannot speak a reply; the Qwen3.5-Omni models handle speech generation. Weights are not published.

MetricValue
Context Length1,000,000 tokens
Max Output131,072 tokens
MultilingualYes

Ready to build with Qwen3.8 Omni Flash?

Try Qwen3.8 Omni Flash in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.

curl -sSf -X POST https://hub.oxen.ai/api/ai/chat/completions \
    -H "Content-Type: application/json" \
    -H "Authorization: Bearer $OXEN_API_KEY" \
    -d '{
  "model": "qwen3-8-omni-flash",
  "messages": [
    {
      "role": "user",
      "content": "Try sending a message."
    }
  ]
}'

API endpoint

See the API reference for request and response formats.

POSThttps://hub.oxen.ai/api/ai/chat/completions