Qwen3-235B-A22B-Thinking Benchmark Details

Qwen3-235B-A22B-Thinking currently shows benchmark results led by MMLU-Pro (37 / 175, score 84.40), AIME2025 (35 / 110, score 92.30), LiveCodeBench (45 / 125, score 74.10).

Benchmark Results

Qwen3-235B-A22B-Thinking

Benchmark Results

Knowledge Exams

2 evaluations

Benchmark / mode

Score

Rank/total

MMLU-Pro

Thinking Mode

84.40

37 / 175

HLE

Thinking Mode

18.20

165 / 235

Scientific Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Thinking Mode

81.10

120 / 253

Algorithmic Coding

1 evaluations

Benchmark / mode

Score

Rank/total

LiveCodeBench

Thinking Mode

74.10

45 / 125

Mathematics

3 evaluations

Benchmark / mode

Score

Rank/total

AIME2025

Thinking Mode

92.30

35 / 110

IMO-ProofBench

Thinking Mode

33.30

6 / 16

IMO-ProofBench Advanced

Thinking Mode

5.20

18 / 24

Writing

1 evaluations

Benchmark / mode

Score

Rank/total

Creative Writing

Standard Mode

1365.80

71 / 110

Cross-capability Suites

1 evaluations

Benchmark / mode

Score

Rank/total

LiveBench

Thinking Mode

52.97

88 / 117

Compare with other models