Qwen 27B Unfiltered
The thinker. Dense 27B, refusals removed.
What it is.What it costs.
Qwen's 27B dense model with the refusal direction ablated. Strong reasoning, code and long-form writing, tool calls, and it answers the question you asked instead of the one it was comfortable with.
Billed per token on the API, always. Add :online and a search call is billed too. Plans make the chat app free; they do not change these rates.
What it can do.Then call it.
Click a capability to see the request parameter behind it. The call on the right works with any OpenAI SDK: point it at our base URL and swap the model id.
OpenAI-compatible chat completions, with streaming.
"stream": trueFunction calling in the OpenAI format: send tools, get tool_calls back.
"tools": [{ "type": "function", "function": { "name": "get_weather", ... } }]JSON mode: ask for a JSON object and get one.
"response_format": { "type": "json_object" }Works through hard problems step by step before it answers.
Append :online and the answer is grounded in live web results, with sources.
"model": "venym/qwen-27b:online"curl https://venym.io/v1/chat/completions \
-H "Authorization: Bearer $VENYM_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "venym/qwen-27b",
"messages": [{"role": "user", "content": "Hello."}]
}'What Qwen costsat your volume.
Drag the sliders. The price comes from the same table that bills you.
A plan ($12.00/mo for Pro) makes the chat app free. The API stays per token: this bill does not change.