How to Use Qwen3.7 Flash on Composite

At a glance

Qwen3.7 Flash is a vision-language reasoning model from Alibaba. Qwen describes it as suited for multimodal agents, visual coding, search and computer interaction, with strengths in object recognition and spatial understanding. It has a one-million-token window and can write replies of up to 65,536 tokens.

What to use it for

Its strengths are on the vision side. Good tests are a screenshot of an interface you want rebuilt, a photo you want described for a scene, or a diagram you want explained. Through the API, send images as image_url content parts; video works the same way where your client supports it.

For plain text chat it accepts temperature, top P and presence penalty, but not min P, top K, or repetition and frequency penalties. Presets that rely on those settings will run, but those settings have no effect on this model.

Long prompts on the free ID

On the paid ID, qwen/qwen3.7-flash, the price per token goes up once a prompt passes 32,000 tokens, and again past 256,000. The free ID does not work that way: every call counts as one of your 80 daily requests, however long the prompt is. That makes the free ID a reasonable place to try long documents or long chat histories. The model page shows the current paid tiers.

Reasoning

The model accepts a reasoning setting and can return its reasoning, but the catalog lists no effort levels for it, so a bracket suffix such as [high] may have no effect. Turn reasoning on or off in the request body or your app's settings instead.

Connect it to your chat app

1. Create an API key on the API Keys page.
2. In SillyTavern, ChubAI, Agnai or RisuAI, choose an OpenAI-compatible provider and set the base URL to https://composite.lucidity.sh/v1. JanitorAI wants the full URL: https://composite.lucidity.sh/v1/chat/completions.
3. Paste the key, then type qwen/qwen3.7-flash:free as the model name.
4. Send a short message to check the connection before you load a long chat.

Step-by-step setup for each app is in the JanitorAI and SillyTavern guides, and the docs hub covers the others.

From code, it is a normal OpenAI-compatible request:

curl https://composite.lucidity.sh/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "qwen/qwen3.7-flash:free", "messages": [{"role": "user", "content": "Hello"}]}'

Free access (as of October 21, 2026)

On October 21, 2026, this model is on Composite's free list as qwen/qwen3.7-flash:free. The plain qwen/qwen3.7-flash stays paid. Every account shares 80 requests a day across all of the free models, separate from Synth's count, and the count resets at midnight UTC. Credits and plans do not raise it.

The free list changes. Check the free models on the Models page for the current set, and the features guide for the rules. A :free id that is no longer on the list returns a model-not-found error and charges nothing. When the 80 requests are used up, requests stop with an error until the reset; Composite never charges your credits to continue. You can switch to qwen/qwen3.7-flash on credits or a plan, or to Synth.

More

• Model page with live pricing and rankings
• Model modifiers, including reasoning effort suffixes
• Best LLMs for Roleplay in 2026
• All model guides