Qwen 3 235B-A22B

Qwen 3 235B-A22B is Alibaba's largest and most capable model, using a mixture-of-experts architecture with 235 billion total parameters and 22 billion active parameters per token. This design delivers frontier-level performance while keeping per-token compute costs manageable. The model features a 128K context window and excels across code generation, mathematical reasoning, multilingual tasks, tool use, and creative writing. It represents the top of the Qwen 3 lineup, competing with the strongest open-weight models available.