Qwen: Qwen3 235B A22B Thinking 2507

Live

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...

Start generating See pricing

FreeInd is an independent platform and is not affiliated with, endorsed by, or sponsored by Qwen or the creators or owners of Qwen: Qwen3 235B A22B Thinking 2507. The Qwen name and mark appear solely to identify the integrations available on this platform.

Rate: 7 credits; per generation
Modality: text; Audio generation
Real samples: 0; live outputs from this model
Provider routes: 1; single routing provider

Capabilities

The generation modes and formats Qwen: Qwen3 235B A22B Thinking 2507 supports on this platform.

text_to_text

More from Qwen

Other live Qwen models on the platform.

All Qwen models

Qwen3.8 2.4T A95B7 credits0 samples Qwen3Guard 8B7 credits0 samples Qwen3 Coder 30B7 credits0 samples Qwen: Qwen-Audio-3.0-TTS Flash7 credits0 samples Qwen: Qwen-Audio-3.0-TTS Plus9 credits0 samples Qwen: Qwen Image 313 credits0 samples

Other makers

Continue browsing the catalog by original model maker.

OpenAI(86)Google(38)Meta(9)Black Forest Labs(5)xAI(9)ByteDance(14)Kling AI(3)MiniMax(14)