GROK

Grok 3

Grok / xAI

29scores
29benchmarks
$3 / $15 per 1M tokenscost in/out

Metadata

Grok Closed/API

Aliases: grok-3, x-ai-grok-3, x-ai/grok-3

Benchmark Results

Benchmark Category Rank Score Sampled
APEX-Agents Agentic 37 7.30 2026-05-06
ARC-AGI-1 Agentic 134 5.50 2026-05-05
ARC-AGI-2 Agentic 139 0 2026-05-05
Tau2-Bench Telecom Agentic 180 48.8% 2026-05-11
Terminal-Bench Hard Agentic 198 11.4% 2026-05-11
OpenUGI Alignment 10 63.47 2026-05-06
LiveCodeBench Coding 85 52.901% 2026-05-28
SciCode Coding 157 36.8% 2026-05-11
K-12EduBench Education 16 63.91 2026-05-27
CorpFin v2 Finance 52 59.713% 2026-05-28
FinanceArena Finance 7 44.6 2026-05-27
TaxEval v2 Finance 4 75.879% 2026-05-28
MedQA Healthcare 63 83.85% 2026-04-16
HUMAINE Human Preference 11 3.67 2026-05-06
AIIQ Composite IQ Intelligence 39 85 2026-05-12
Artificial Analysis Intelligence Index Intelligence 191 25.17 2026-05-11
GPQA Diamond Intelligence 58 74.242% 2026-05-28
Humanity's Last Exam Intelligence 291 5.1% 2026-05-11
MMLU Pro Intelligence 63 79.949% 2026-05-28
MMLU-Pro Intelligence 117 79.9% 2026-05-11
LegalBench Legal 38 82.595% 2026-05-28
AIME Math 58 58.75% 2026-04-16
AIME 2025 Math 120 58% 2026-05-11
IneqMath Math 34 3.50 2026-05-06
MATH 500 Math 27 89.8% 2026-01-09
MGSM Math 28 91.346% 2026-01-09
Design Arena Multimodal 92 1111 2026-05-06
GPQA Diamond Reasoning 212 69.3% 2026-05-11
CritPt Science 241 0% 2026-05-11