Qwen 3.8 Omni Flash

Qwen 3.8 Omni Flash (qwen/qwen3.8-omni-flash) — the light tier of the Qwen 3.8 omni model: understands text and images, cheap and fast. Thinking is always on (depth not controllable); files are not accepted.

Parameters and capabilities

Parameter	Value
model	qwen/qwen3.8-omni-flash (short alias qwen3.8-omni-flash)
Reasoning	always on, reasoning_effort is not supported
Images on input	yes (image_url, data URLs)
Files on input	no — a file part returns 400
Cache read	$0.02 per 1M tokens

Example

curl https://api.mixen.ai/v1/chat/completions \
  -H "Authorization: Bearer $MIXEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen/qwen3.8-omni-flash",
    "messages": [{
      "role": "user",
      "content": [
        {"type": "text", "text": "What is in this screenshot and what is wrong with it?"},
        {"type": "image_url", "image_url": {"url": "data:image/png;base64,..."}}
      ]
    }]
  }'

Pricing

Per 1M tokens: input $0.188, output $0.588, cache read $0.02. Current prices — catalog.

Strengths and limitations

• A cheap entry into multimodal tasks: image+text at mass-tier text pricing.
• The reasoning depth is not selectable; reasoning tokens are always billed.
• Files (PDF/DOCX) are not accepted — images only.

All models — in the catalog. The general workflow — chat guide.