Grok 4 Fast Benchmark Details

Grok 4 Fast currently shows benchmark results led by SimpleQA (3 / 46, score 95), Fiction.liveBench (3 / 16, score 94.40), AIME2025 (45 / 215, score 92).

Benchmark Results

Grok 4 Fast

Benchmark Results

Knowledge Exams

3 evaluations

Benchmark / mode

Score

Rank/total

HLE

Standard Mode

4.50

506 / 568

HLE

Thinking Mode

292 / 568

HLE

Thinking Mode

19.10

305 / 568

Scientific Reasoning

3 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Standard Mode

60.60

375 / 461

GPQA Diamond

Thinking Mode

85.70

152 / 461

CritPt

Thinking Mode

2.90

119 / 204

Honesty & Factuality

1 evaluations

Benchmark / mode

Score

Rank/total

SimpleQA

Thinking ModeTools

3 / 46

Algorithmic Coding

2 evaluations

Benchmark / mode

Score

Rank/total

LiveCodeBench

Standard Mode

40.10

186 / 251

LiveCodeBench

Thinking Mode

53 / 251

Mathematics

2 evaluations

Benchmark / mode

Score

Rank/total

AIME2025

Standard Mode

41.30

167 / 215

AIME2025

Thinking Mode

45 / 215

Service Workflows

5 evaluations

Benchmark / mode

Score

Rank/total

τ²-Bench - Telecom

Standard ModeTools

63.70

152 / 264

τ²-Bench - Telecom

Thinking ModeTools

65.80

145 / 264

SAGE

Standard ModeTools

15.99

63 / 64

SAGE

Thinking ModeTools

29.76

60 / 64

τ³-Banking

Thinking Level · HighTools

15.72

106 / 167

Instruction Following

2 evaluations

Benchmark / mode

Score

Rank/total

IF Bench

Standard Mode

37.70

234 / 282

IF Bench

Thinking Mode

50.50

155 / 282

Agentic Development

2 evaluations

Benchmark / mode

Score

Rank/total

Terminal Bench Hard

Standard ModeTools

12.10

178 / 244

Terminal Bench Hard

Thinking ModeTools

18.90

150 / 244

Long Reasoning

2 evaluations

Benchmark / mode

Score

Rank/total

Fiction.liveBench

Standard Mode

94.40

3 / 16

AA-LCR

Standard Mode

169 / 174

Visual Understanding

2 evaluations

Benchmark / mode

Score

Rank/total

MMMU-Pro

Standard Mode

48.10

206 / 229

MMMU-Pro

Thinking Mode

61.80

170 / 229

ML Engineering

1 evaluations

Benchmark / mode

Score

Rank/total

WeirdML v2

Standard ModeTools

42.86

45 / 52

Preference Arenas

1 evaluations

Benchmark / mode

Score

Rank/total

Text Arena (Coding)

Standard Mode

1149

35 / 35

Code Generation & Editing

1 evaluations

Benchmark / mode

Score

Rank/total

Vibe Code Bench v1.1

Thinking ModeTools

58 / 60

Clinical Workflows

4 evaluations

Benchmark / mode

Score

Rank/total

MedScribe

Standard ModeTools

79.72

41 / 66

MedScribe

Thinking ModeTools

81.63

38 / 66

MedCode

Standard ModeTools

30.04

60 / 64

MedCode

Thinking ModeTools

37.38

51 / 64

Capability Indices

1 evaluations

Benchmark / mode

Score

Rank/total

ECI

unknown

144.21

80 / 167

Compare with other models