benchgap
xAI · model

Grok 4 benchmark scores

As of 2026-11-21, Grok 4 (xAI) has measured scores on 20 benchmarks and estimated scores on 89 more.

BenchmarkScoreSource
AA-GPQA Diamond87.7%measured
AA-HLE26.7%measured
AA-SciCode42.7%estimated ± 4.2 pp, high confidence
Artificial Analysis Intelligence Index22.5%measured
CritPt2.0%measured
GDPval-AA17.9%estimated ± 8.8 pp, medium confidence
GPQA Diamond89.8%estimated ± 1.8 pp, high confidence
HLE28.7%estimated ± 8.5 pp, medium confidence
IFBench57.1%estimated ± 6.6 pp, medium confidence
AA-LCR68.0%measured
LiveCodeBench v687.9%estimated ± 2.2 pp, medium confidence
AA-Omniscience Accuracy40.5%measured
SciCode43.1%estimated ± 2.9 pp, low confidence
BrowseComp63.3%estimated ± 6.6 pp, medium confidence
HLE w/ tools25.1%estimated ± 5.5 pp, low confidence
IFEval91.6%estimated ± 2.8 pp, high confidence
LongBench v257.5%estimated ± 5.5 pp, medium confidence
τ²-bench results74.9%measured
AA Coding Index34.7%estimated ± 5.6 pp, medium confidence
AA-MMMU-Pro68.8%measured
APEX-Agents-AA12.2%estimated ± 11.2 pp, low confidence
BioMysteryBench (human-difficult)25.6%estimated ± 6.8 pp, low confidence
SWE-bench Verified70.5%estimated ± 4.4 pp, high confidence
BFCL v458.6%estimated ± 8.5 pp, low confidence
CyberGym34.8%estimated ± 9.3 pp, low confidence
JobBench18.7%estimated ± 10.3 pp, low confidence
SWE-bench Pro57.2%estimated ± 4.7 pp, high confidence
WideResearch63.4%estimated ± 5.9 pp, low confidence
AA Agentic Index12.3%estimated ± 6.8 pp, medium confidence
MMLU-Pro85.0%estimated ± 4.8 pp, high confidence
AA-IFBench53.7%measured
FailureSensorIQ60.4%estimated ± 1.8 pp, low confidence
GPQA89.7%estimated ± 4.5 pp, high confidence
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond88.1%measured
Vals LiveCodeBench83.2%measured
Vals MMLU-Pro85.3%measured
MMMLU87.2%estimated ± 0.7 pp, medium confidence
MMMU79.3%estimated ± 2.7 pp, high confidence
AA MMLU-Pro86.6%measured
Gert Labs42.3%measured
ARC-AGI-183.7%estimated ± 6.5 pp, medium confidence
ARC-AGI-251.2%estimated ± 13.8 pp, low confidence
CursorBench 3.131.3%estimated ± 5.8 pp, low confidence
CursorBench 3.259.1%estimated ± 7.2 pp, low confidence
FrontierCode 1.1 Main9.8%estimated ± 5.2 pp, low confidence
Mystery Game Puzzles6.9%estimated ± 2.2 pp, low confidence
OfficeQA Pro50.3%estimated ± 6.8 pp, low confidence
OSWorld-Verified56.8%estimated ± 5.5 pp, medium confidence
ParseBench-152.7%estimated ± 6.2 pp, medium confidence
Vals SWE-bench57.8%measured
VulcanBench v382.5%estimated ± 5.6 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)52.7%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)57.5%estimated ± 1.1 pp, low confidence
AA ITBench36.5%estimated ± 2.8 pp, high confidence
Chartography (tools)76.1%estimated ± 4.1 pp, low confidence
CursorBench 4.040.0%estimated ± 4.5 pp, high confidence
HLE w/o tools31.3%estimated ± 4.9 pp, high confidence
OSWorld 2.00.0%estimated ± 5.9 pp, low confidence
ARC-AGI-30.0%estimated ± 8.6 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
SkillsBench official leaderboard (BenchFlow)30.2%estimated ± 9.5 pp, low confidence
BinaryAudit8.8%measured
CompileBench71.1%measured
OTelBench4.3%measured
Vibe Code Bench27.7%estimated ± 6.5 pp, low confidence
ParseBench-256.4%estimated ± 6.2 pp, medium confidence
CharXiv76.8%estimated ± 6.3 pp, low confidence
CharXiv w/o tools76.1%estimated ± 2.5 pp, medium confidence
AI2D_TEST88.1%estimated ± 2.0 pp, medium confidence
C-Eval92.3%estimated ± 1.0 pp, medium confidence
Claw-Eval56.5%estimated ± 6.8 pp, low confidence
CountBench90.7%estimated ± 2.9 pp, low confidence
DynaMath52.6%estimated ± 3.0 pp, low confidence
ERQA42.3%estimated ± 3.1 pp, medium confidence
MathVision75.8%estimated ± 3.1 pp, medium confidence
MCP Atlas38.7%estimated ± 14.6 pp, low confidence
MMLU-Redux96.0%estimated ± 0.8 pp, medium confidence
MMMU-Pro70.3%estimated ± 1.5 pp, high confidence
NL2Repo36.8%estimated ± 4.7 pp, medium confidence
PostTrainBench v10.0%estimated ± 3.6 pp, medium confidence
RealWorldQA78.9%estimated ± 2.9 pp, low confidence
ScreenSpot Pro70.5%estimated ± 10.8 pp, low confidence
SuperGPQA82.1%estimated ± 9.3 pp, medium confidence
Toolathlon36.6%estimated ± 7.2 pp, medium confidence
V*34.5%estimated ± 7.5 pp, low confidence
Video-MME (w/o subtitle)79.0%estimated ± 1.7 pp, medium confidence
Video-MME (with subtitle)80.9%estimated ± 3.7 pp, low confidence
AA LiveCodeBench81.9%measured
HealthBench Hard15.3%estimated ± 8.6 pp, low confidence
MedXpertQA (MM)42.5%estimated ± 6.5 pp, low confidence
MMLU-Pro (Arcee)84.6%estimated ± 4.0 pp, medium confidence
React Native Evals72.6%measured
ResearchClawBench14.7%estimated ± 2.2 pp, medium confidence
BioMysteryBench (human-solvable)81.7%estimated ± 0.6 pp, low confidence
HealthBench (raw)35.4%estimated ± 6.0 pp, low confidence
HealthBench (length-adjusted)48.5%estimated ± 3.4 pp, low confidence
HealthBench Professional56.7%estimated ± 4.2 pp, low confidence
HealthBench Professional (raw)46.8%estimated ± 4.5 pp, medium confidence
HLE-Verified15.3%estimated ± 2.0 pp, low confidence
LABBench280.5%estimated ± 1.8 pp, low confidence
STEMBench2.5%estimated ± 0.2 pp, medium confidence
SimpleQA46.3%estimated ± 4.3 pp, low confidence
BabyVision w/ Python83.4%estimated ± 4.0 pp, low confidence
OpenHarmony Bench v1.047.3%estimated ± 3.2 pp, medium confidence
MRCRv264.2%estimated ± 11.3 pp, low confidence
MMLongBench-Doc60.8%estimated ± 2.6 pp, low confidence
BioBasicsBench33.9%estimated ± 2.8 pp, low confidence
MMMU-Pro w/ Python72.6%estimated ± 2.0 pp, high confidence