TAU2
Measured May 29, 2026Source
Score
0.33
Models15 0 
Qwen3 235B A22B 2507 Instruct
阿里巴巴阿里巴巴China
Qwen3 235B A22B is a large-scale Mixture-of-Experts (MoE) language model from Alibaba's Qwen series. It features 235 billion total parameters with 22 billion activated per token, designed for strong instruction following, complex reasoning, and multilingual tasks.
ReasoningCodingLong context
Input / 1M tokens
$0.20Output / 1M tokens
$0.825Output tokens/s
66.15First-token seconds
1.27sSupported plans
0Benchmark history
Evaluations
Terminalbench Hard
Measured May 29, 2026Source
Score
0.15
Lcr
Measured May 29, 2026Source
Score
0.31
Ifbench
Measured May 29, 2026Source
Score
0.46
Aime 25
Measured May 29, 2026Source
Score
0.72
Aime
Measured May 29, 2026Source
Score
0.72
Math 500
Measured May 29, 2026Source
Score
0.98
Scicode
Measured May 29, 2026Source
Score
0.36
Livecodebench
Measured May 29, 2026Source
Score
0.52
Hle
Measured May 29, 2026Source
Score
0.11
Gpqa
Measured May 29, 2026Source
Score
0.75
Mmlu Pro
Measured May 29, 2026Source
Score
0.83
Artificial Analysis Math Index
Measured May 29, 2026Source
Score
71.7
Artificial Analysis Coding Index
Measured May 29, 2026Source
Score
22.1
Artificial Analysis Intelligence Index
Measured May 29, 2026Source
Score
25
Plan availability
Products and plans that support this model
No products or plans have been linked to this model yet.
User ratings
Loading ratings...
Discussion
Thinking... Make sure you are connected to GitHub server