Qwen3 235B A22B Instruct 2507
Qwen🇨🇳
qwen/qwen3-235b-a22b-2507A July 2025 refreshed release of Qwen3 235B-A22B, Alibaba's large MoE model, with improved instruction following and general capability over the original version. Suited for demanding general-purpose, coding, and agentic workloads.
Cost rate
0.1x
Context
262K
Released
Jul 21, 2025
Input
Text
Output
Text
Support
ReasoningTool callingStructured outputs
Best at
The categories where this model ranks highest.
Medicine & Healthcare
#100 · top 26%
Business, Management, & Finance
#112 · top 28%
Software & IT Services
#117 · top 28%
Life, Physical, & Social Science
#120 · top 29%
Performance
Median latency and throughput measured across recent requests.
Throughput63 tok/s
Latency