Qwen3 235B A22B
Qwen
qwen/qwen3-235b-a22bQwen3 235B-A22B is a large mixture-of-experts model (235B total, 22B active parameters) from Alibaba's Qwen3 family, offering strong multilingual reasoning, coding, and agentic capabilities. Its MoE design gives near-flagship quality at lower inference cost than an equivalent dense model.
Cost rate
0.4x
Context
131K
Released
Apr 28, 2025
Input
Text
Output
Text
Support
ReasoningTool callingStructured outputs
Best at
The categories where this model ranks highest.
Medicine & Healthcare
#86 · top 24%
Business, Management, & Finance
#94 · top 24%
Software & IT Services
#101 · top 26%
Life, Physical, & Social Science
#102 · top 26%
Performance
Median latency and throughput measured across recent requests.
Throughput59 tok/s
Latency