AI-SOVEREIGN
Home / AI Model APIs / Qwen3 235B A22B Thinking 2507
QWAI Model APIs

Qwen3 235B A22B Thinking 2507

by Qwen

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...

Input price$0.45
Output price$3.50
Context128K tokens
Released25 Jul 2025
Modalitytext->text

More from Qwen