Grok 4 Benchmark Details

Grok 4 currently shows benchmark results led by MMLU-Pro (15 / 175, score 87), IOI 2024 (Vals v1) (1 / 10, score 23.20), IOI 2025 (Vals v1) (1 / 9, score 29.20).

Benchmark Results

Grok 4

Benchmark Results

Knowledge Exams

4 evaluations

Benchmark / mode

Score

Rank/total

MMLU-Pro

Thinking Enabled

15 / 175

HLE

Thinking Enabled

25.40

133 / 235

HLE

Thinking EnabledTools

38.60

86 / 235

HLE

Thinking EnabledToolsInternet

38.60

86 / 235

Abstract Generalization

2 evaluations

Benchmark / mode

Score

Rank/total

ARC-AGI-1

Thinking Enabled

66.70

117 / 193

ARC-AGI-2

Thinking Enabled

15.90

120 / 181

Scientific Reasoning

2 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Thinking Enabled

70 / 253

CritPt

Thinking Enabled

131 / 204

Repository Engineering

1 evaluations

Benchmark / mode

Score

Rank/total

SWE-bench Verified

Thinking Enabled

58.60

85 / 116

Mathematics

7 evaluations

Benchmark / mode

Score

Rank/total

AIME2025

Thinking Enabled

91.70

38 / 110

AIME2025

Thinking EnabledTools

98.80

14 / 110

IMO-ProofBench

Thinking Enabled

46.70

4 / 16

IMO-ProofBench

Deep

23.30

10 / 16

IMO-ProofBench Advanced

Thinking Enabled

18.60

12 / 24

FrontierMath

Standard Mode

12.10

22 / 60

FrontierMath - Tier 4

Standard Mode

2.10

56 / 80

Agentic Development

3 evaluations

Benchmark / mode

Score

Rank/total

Aider-Polyglot

Thinking Level · High

79.60

7 / 59

Terminal-Bench

Thinking EnabledTools

13 / 35

Terminal Bench Hard

Thinking EnabledTools

37.90

59 / 244

Commonsense

1 evaluations

Benchmark / mode

Score

Rank/total

SimpleBench

Thinking Enabled

60.50

36 / 96

Algorithmic Coding

2 evaluations

Benchmark / mode

Score

Rank/total

IOI 2025 (Vals v1)

Thinking EnabledTools

29.20

1 / 9

IOI 2024 (Vals v1)

Thinking EnabledTools

23.20

1 / 10

Cross-capability Suites

1 evaluations

Benchmark / mode

Score

Rank/total

LiveBench

Standard Mode

62.02

61 / 117

Visual Understanding

2 evaluations

Benchmark / mode

Score

Rank/total

MMMU-Pro

Thinking Enabled

68.80

141 / 229

GeoBench ACW

Standard Mode

20 / 20

ML Engineering

1 evaluations

Benchmark / mode

Score

Rank/total

WeirdML v2

Standard ModeTools

45.73

41 / 52

Long Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

Fiction.liveBench

Standard Mode

94.40

3 / 16

Logical Planning

1 evaluations

Benchmark / mode

Score

Rank/total

BALROG

Standard ModeTools

43.60

4 / 12

Capability Frontier Metrics

1 evaluations

Benchmark / mode

Score

Rank/total

METR Time Horizons v1.1

Standard ModeTools

110.07

13 / 22

Capability Indices

1 evaluations

Benchmark / mode

Score

Rank/total

ECI

unknown

146.46

66 / 167

Compare with other models