Model
Qwen3.8 27B Uncensored
Qwen3.8 27B Uncensored is a model for coding and chat with the refusal behaviour removed.
Specs
- Model id
- qwen3.8-27b-uncensored
- Context window
- 262,144 tokens
- Input
- $1.00 per million tokens
- Cached input
- $0.10 per million tokens
- Output
- $3.00 per million tokens
- Reasoning
- Returned in reasoning_content, apart from the reply
- API
- OpenAI-compatible chat completions, completions and responses; Anthropic-compatible messages
- Privacy
- Prompts and completions are never saved or logged.
Cached-input pricing applies to the API and coding tools; messages in the web chat pay the full input price.
What it does
No refusals
It answers the question you asked. Fiction, research, security testing with permission and difficult topics are all fine.
262K context
Enough room for a whole repository, a long document or a long conversation.
Dedicated GPUs
Served on dedicated GPUs that run only this model.
Tool calls
Replies stream as they are generated, and tool calls work in coding agents and in your own code.
Use it
Call it as qwen3.8-27b-uncensored through the API, or connect a coding agent with one of the setup guides. Every price is on the pricing page.