Inference
EN
xAI

Grok 4

Grok 4 is a standard model from xAI. 21 benchmarks count toward its index, in 7 categories.

  • Closed weights
  • Standard
  • Input: text
  • Released 2025-07
RankedEnough results for an index.
IndexAll categories, out of 100.50.0±4.0Median 51.0
CoverageIndex weight that has results.95%
SpeedOutput tokens per second.78/sMedian 67/s
First tokenSeconds to the first output token.11.6 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.—Median $0.60
Output / 1MUS dollars per 1M output tokens.—Median $2.55
ContextMaximum tokens in one request.128KMedian 262K
EloLMArena rating and rank.1411±4 #141Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.4 counted48.0147 / 218
CodingCode writing and repair.4 counted50.9132 / 209
ReasoningLogic problems and puzzles.2 counted47.3135 / 227
MultimodalTasks with images and text.1 counted49.6—
KnowledgeFacts and expert knowledge.4 counted55.887 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted39.2—
MathMath problems.5 counted50.585 / 168

Results

21 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MATH 500Math96.2% ±0.92 / 59
GPQA Diamond · Vals AIKnowledge88.1% ±1.635 / 135
Artificial Analysis GPQA DiamondKnowledge87.7%55 / 199
IOI v1Coding26.2% ±6.817 / 61
AIMEMath90.6% ±1.827 / 95
Artificial Analysis Omniscience AccuracyKnowledge40.5%48 / 164
GPQA diamond · Epoch AIKnowledge87.0% ±2.064 / 217
Artificial Analysis MATH-500Math99.0%3 / 9
LiveCodeBenchCoding83.2% ±1.049 / 142
OTIS Mock AIME 2024-2025Math84.0% ±5.077 / 200
Artificial Analysis Humanity's Last ExamKnowledge26.7%95 / 222
MMLU ProKnowledge85.3% ±0.358 / 135
FrontierMath-2025-02-28-PrivateMath19.7% ±2.331 / 69
Critical Physics TasksReasoning2.0%105 / 214
Artificial Analysis IFBenchInstruction53.7%73 / 146
τ²-Bench Tool-Agent-User EvaluationAgentic74.9%75 / 149
Artificial Analysis MMLU-ProKnowledge86.6%8 / 15
Terminal-Bench 1.0Agentic38.8% ±5.426 / 47
Artificial Analysis Long Context ReasoningReasoning68.0%125 / 217
MMMU ProMultimodal76.3% ±1.054 / 92
Gert Labs Composite Game BenchmarkAgentic42.3%33 / 52
FrontierMath-Tier-4-2025-07-01-PrivateMath2.1% ±2.135 / 55
Artificial Analysis MMMU-ProMultimodal68.8%79 / 113
Terminal-Bench 2.0 · Vals AIAgentic28.1% ±4.847 / 66
React Native EvalsCoding72.6%12 / 16
Artificial Analysis LiveCodeBenchCoding81.9%10 / 12
SWE-benchCoding57.8% ±2.278 / 87
FrontierMath-Tier-4-2025-07-01-PublicMath0.0%3 / 36
CaseLaw v2Legal65.8% ±0.29 / 53
CorpFin v2Finance66.0% ±0.923 / 131
MedQAHealthcare92.5% ±0.230 / 95
Finance Agent v1.1Finance53.5% ±2.919 / 51
LegalBenchLegal83.2% ±0.557 / 149
MGSMMultilingual90.9% ±1.831 / 74
MedScribeHealthcare78.2% ±2.166 / 109
MedCodeHealthcare38.1% ±2.272 / 106
TaxEval v2Finance65.1% ±1.0116 / 143
MortgageTaxFinance44.5% ±1.087 / 97
Artificial Analysis AIME 2025Math92.7%—
Artificial Analysis Global-MMLU-LiteMultilingual89.5%—

Sources

BenchLMCC BY-NC 4.0Vals AINo licence statedEpoch AICC BY

History

5 releases
Index51.0 → 50.0
23 Oct 202612 Nov 2026

Questions

How good is Grok 4?
Its index is 50.0 out of 100. That is place 150 of 283 ranked models.
What is Grok 4 best at?
Knowledge: 55.8 out of 100, place 87 of 268.
How fast is Grok 4?
78 output tokens per second, 11.6 s to the first token.
What is the context window of Grok 4?
128K tokens.
Is Grok 4 open source?
No. Its weights are not published.
When was Grok 4 released?
July 2025.

Badge

Grok 4 on Inference360: place 150 of 283, index 50.0
HTML
<a href="https://inference360.ai/models/grok-4"><img src="https://inference360.ai/badges/grok-4.svg" alt="Grok 4 on Inference360: place 150 of 283, index 50.0" height="20"></a>
Markdown
[![Grok 4 on Inference360: place 150 of 283, index 50.0](https://inference360.ai/badges/grok-4.svg)](https://inference360.ai/models/grok-4)

Compare with

GLM-4.750.0MiniMax M2.550.2Gemini 3.1 Flash Lite Preview49.6Qwen3.5-35B-A3B49.4Grok Code Fast 141.2

More from xAI

All models
Grok Code Fast 141.2Grok 4 Fast (Reasoning)45.3Grok 4.1UnrankedGrok 4.1 Fast (Reasoning)47.3Grok 4.1 Fast33.3