Claude Opus 4 Benchmark Details

Claude Opus 4 currently shows benchmark results led by MATH-500 (3 / 44, score 98.20), MMLU Pro (23 / 124, score 85), Simple Bench (7 / 27, score 58.80).

Benchmark Results

Claude Opus 4

Benchmark Results

综合评估

5 evaluations

Benchmark / mode

Score

Rank/total

MMLU Pro

Standard Mode

23 / 124

GPQA Diamond

Standard Mode

79.60

76 / 175

ARC-AGI

Standard Mode

35.70

48 / 65

HLE

Standard Mode

10.70

121 / 149

ARC-AGI-2

Standard Mode

8.60

38 / 58

编程与软件工程

2 evaluations

Benchmark / mode

Score

Rank/total

SWE-bench Verified

Standard Mode

72.50

43 / 103

LiveCodeBench

Standard Mode

56.60

74 / 118

数学推理

9 evaluations

Benchmark / mode

Score

Rank/total

MATH-500

Standard Mode

98.20

3 / 44

AIME 2024

Standard Mode

35 / 62

AIME2025

Standard Mode

75.50

65 / 106

FrontierMath

Standard Mode

4.50

39 / 60

FrontierMath

Thinking Enabled

4.10

41 / 60

FrontierMath - Tier 4

Standard Mode

72 / 80

FrontierMath - Tier 4

Thinking Enabled

4.20

40 / 80

FrontierMath - Tier 4

32K

4.20

40 / 80

IMO-ProofBench

Thinking Enabled

2.90

16 / 16

写作和创作

1 evaluations

Benchmark / mode

Score

Rank/total

Creative Writing

Standard Mode

83.75

13 / 23

常识推理

1 evaluations

Benchmark / mode

Score

Rank/total

Simple Bench

Thinking Enabled

58.80

7 / 27

Agent能力评测

3 evaluations

Benchmark / mode

Score

Rank/total

τ²-Bench

Thinking EnabledTools

72.50

22 / 40

Aider-Polyglot

Standard Mode

70.10

11 / 26

Aider-Polyglot

Thinking Enabled

8 / 26

Compare with other models