Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...
Modalities
Text
In / out price
₩404 / ₩4,037
1M
Context
13.11万
リリース日
2025/07/25
稼働率データはありません。