benchgap
xAI · model

Grok 4.5 benchmark scores

As of 2026-11-22, Grok 4.5 (xAI) has measured scores on 39 benchmarks and estimated scores on 138 more.

BenchmarkScoreSource
AA-GPQA Diamond93.1%measured
AA-HLE42.7%measured
AA-SciCode55.0%measured
AIME2696.0%estimated ± 1.2 pp, medium confidence
Artificial Analysis Intelligence Index38.8%measured
CritPt15.4%measured
GDPval-AA44.2%measured
GPQA Diamond89.9%estimated ± 1.4 pp, high confidence
HLE54.3%estimated ± 2.2 pp, high confidence
AA-LCR79.3%measured
LiveCodeBench v689.8%estimated ± 2.2 pp, low confidence
AA-Omniscience Accuracy51.6%measured
SciCode51.5%estimated ± 2.9 pp, medium confidence
Terminal-Bench 2.183.3%measured
BrowseComp87.2%estimated ± 4.7 pp, high confidence
HLE w/ tools55.2%estimated ± 5.3 pp, medium confidence
LongBench v265.1%estimated ± 5.5 pp, medium confidence
AA Coding Index72.5%measured
AA-MMMU-Pro80.4%measured
APEX-Agents36.4%estimated ± 9.4 pp, low confidence
APEX-Agents-AA27.0%estimated ± 9.3 pp, medium confidence
BioMysteryBench (human-difficult)46.7%estimated ± 6.8 pp, low confidence
DeepSearchQA92.1%estimated ± 4.5 pp, high confidence
SWE-bench Verified84.5%estimated ± 3.6 pp, high confidence
AutomationBench30.2%estimated ± 7.1 pp, low confidence
BFCL v476.2%estimated ± 2.7 pp, medium confidence
CyberGym81.9%estimated ± 3.1 pp, low confidence
JobBench46.1%estimated ± 8.7 pp, medium confidence
SWE-bench Pro64.7%measured
WideResearch79.6%estimated ± 3.3 pp, high confidence
AA Agentic Index42.1%measured
MMLU-Pro87.2%estimated ± 4.8 pp, high confidence
FailureSensorIQ67.3%estimated ± 1.7 pp, high confidence
GPQA93.1%estimated ± 0.9 pp, high confidence
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond92.9%measured
Vals LiveCodeBench87.4%measured
Vals MMLU-Pro89.2%measured
MMMLU90.3%estimated ± 0.7 pp, low confidence
MMMU86.7%estimated ± 2.7 pp, medium confidence
Chess Puzzles35.0%measured
AA MMLU-Pro89.0%estimated ± 0.7 pp, medium confidence
Gert Labs47.2%measured
AA Harvey LAB v1.092.6%estimated ± 1.2 pp, medium confidence
AA Tau3 Banking42.1%measured
AA Terminal-Bench 2.181.6%measured
ApprenticeBench6.2%estimated ± 10.4 pp, low confidence
ARC-AGI-185.7%measured
ARC-AGI-252.6%measured
BabaIsBench75.0%measured
CursorBench 3.156.3%estimated ± 3.4 pp, medium confidence
CursorBench 3.266.7%measured
EBR-bench8.2%estimated ± 9.3 pp, medium confidence
Terminal-Bench 3.015.7%measured
FrontierCode 1.1 Main49.9%estimated ± 1.6 pp, medium confidence
FrontierMath v2 (Tier 4)24.4%measured
FrontierMath v2 (Tiers 1-3)57.2%measured
FrontierSWE v223.7%estimated ± 8.3 pp, medium confidence
Furniture Assembly22.5%measured
MaxxBench49.1%estimated ± 6.6 pp, low confidence
MazeBench0.5%estimated ± 2.7 pp, high confidence
MirrorCode32.2%measured
Mystery Game Puzzles26.0%estimated ± 10.3 pp, low confidence
NYT Connections79.9%measured
OfficeQA Pro59.5%estimated ± 6.8 pp, medium confidence
OSWorld-Verified78.8%estimated ± 8.4 pp, low confidence
ParseBench-165.8%estimated ± 6.2 pp, medium confidence
Terminal-Bench 4.112.4%measured
Terminal-Bench Hard55.9%estimated ± 2.8 pp, medium confidence
Vals SWE-bench86.6%measured
Terminal-Bench 2.1 (Vals AI)67.8%measured
VulcanBench v389.9%measured
CWE-Bench-AA50.7%estimated ± 8.1 pp, medium confidence
AA Cyber Index25.5%estimated ± 9.2 pp, low confidence
DeepsecBench-AA16.1%estimated ± 8.2 pp, medium confidence
HealthBench Professional (length-adjusted, Artificial Analysis)57.7%estimated ± 4.3 pp, medium confidence
HealthBench Professional (raw, Artificial Analysis)62.0%estimated ± 0.8 pp, low confidence
AA ITBench46.3%estimated ± 4.0 pp, high confidence
Bug Hunt Bench7.0%estimated ± 6.3 pp, low confidence
Chartography (tools)83.6%estimated ± 4.1 pp, low confidence
CursorBench 4.036.6%estimated ± 2.0 pp, high confidence
CWE-bench v149.0%estimated ± 5.7 pp, medium confidence
DeepSWE53.0%measured
DRACO82.4%estimated ± 2.0 pp, medium confidence
FrontierCode 2.025.4%estimated ± 5.4 pp, low confidence
FrontierPhysics0.0%estimated ± 4.4 pp, high confidence
HLE w/o tools47.8%estimated ± 4.9 pp, high confidence
OSWorld 2.157.6%estimated ± 3.6 pp, high confidence
OSWorld 2.1 Offline42.1%estimated ± 1.9 pp, low confidence
PostTrainBench v1.123.5%measured
ProgramBench0.0%estimated ± 4.8 pp, low confidence
SpireClimb5.8%estimated ± 7.2 pp, medium confidence
StealthMark45.2%estimated ± 6.5 pp, medium confidence
SWE Multilingual78.0%measured
Terminal-Bench 4.019.9%estimated ± 4.5 pp, high confidence
Terminal-Bench-Science 0.10.0%estimated ± 10.7 pp, low confidence
Toolathlon-Verified69.5%estimated ± 2.3 pp, medium confidence
VulcanBench Frontier v488.4%estimated ± 2.3 pp, medium confidence
AA AutomationBench52.7%estimated ± 9.9 pp, low confidence
GDP.pdf19.1%estimated ± 4.2 pp, high confidence
AA Terminal-Bench 4.012.0%estimated ± 6.2 pp, medium confidence
ARC-AGI-30.3%measured
ARC-AGI-3 Provider Adapter1.6%estimated ± 4.1 pp, medium confidence
cursorBench4144.8%estimated ± 3.7 pp, high confidence
CWE-bench45.2%estimated ± 6.2 pp, medium confidence
MKA35.0%estimated ± 4.6 pp, high confidence
SkillsBench official leaderboard (BenchFlow)54.3%estimated ± 6.1 pp, medium confidence
BinaryAudit35.8%estimated ± 12.1 pp, low confidence
OTelBench22.5%estimated ± 6.6 pp, medium confidence
Vibe Code Bench61.2%estimated ± 10.0 pp, low confidence
FACTS Multimodal45.1%estimated ± 4.0 pp, high confidence
Hermes Bench67.9%estimated ± 2.9 pp, high confidence
Hermes Index30.9%estimated ± 5.7 pp, low confidence
ParseBench-272.4%estimated ± 6.2 pp, medium confidence
CharXiv84.3%estimated ± 6.3 pp, low confidence
CharXiv w/o tools83.5%estimated ± 2.5 pp, high confidence
ExploitBench32.4%estimated ± 6.5 pp, low confidence
ExploitGym3.1%estimated ± 8.3 pp, low confidence
AI2D_TEST94.1%estimated ± 2.0 pp, low confidence
C-Eval97.2%estimated ± 1.0 pp, low confidence
CountBench98.4%estimated ± 2.9 pp, low confidence
DynaMath85.5%estimated ± 3.0 pp, low confidence
ERQA68.7%estimated ± 3.1 pp, high confidence
HMMT Feb 202692.7%estimated ± 1.5 pp, medium confidence
HMMT Nov 202594.9%estimated ± 1.5 pp, medium confidence
INCLUDE85.9%estimated ± 2.0 pp, medium confidence
MathVision90.9%estimated ± 3.1 pp, high confidence
MCP Atlas77.9%estimated ± 6.0 pp, low confidence
MMAnswerBench88.4%estimated ± 2.4 pp, medium confidence
MMLU-ProX85.9%estimated ± 1.3 pp, high confidence
MMLU-Redux97.4%estimated ± 0.8 pp, low confidence
MMMU-Pro80.9%estimated ± 1.5 pp, high confidence
NL2Repo50.6%estimated ± 4.7 pp, high confidence
PostTrainBench v125.6%estimated ± 1.3 pp, medium confidence
QwenClawBench93.2%estimated ± 6.0 pp, low confidence
RealWorldQA86.4%estimated ± 2.9 pp, medium confidence
ScreenSpot Pro77.7%estimated ± 10.8 pp, low confidence
SuperGPQA100.0%estimated ± 9.3 pp, low confidence
Terminal-Bench 2.071.6%estimated ± 5.8 pp, medium confidence
Toolathlon55.9%estimated ± 6.3 pp, medium confidence
V*100.0%estimated ± 7.5 pp, low confidence
Video-MME (w/o subtitle)87.0%estimated ± 1.7 pp, low confidence
Video-MME (with subtitle)88.5%estimated ± 3.7 pp, medium confidence
AA LiveCodeBench95.6%estimated ± 2.6 pp, medium confidence
HealthBench Hard31.4%estimated ± 8.6 pp, low confidence
LiveCodeBench Pro87.3%estimated ± 6.0 pp, low confidence
MedXpertQA (MM)77.1%estimated ± 6.5 pp, low confidence
React Native Evals82.3%estimated ± 3.2 pp, high confidence
ResearchClawBench20.4%estimated ± 1.9 pp, high confidence
PostTrainBench22.6%measured
DAYJOB1.4%estimated ± 2.7 pp, medium confidence
DAYJOB Healthcare0.0%estimated ± 4.0 pp, medium confidence
Finance Agent v255.6%estimated ± 2.4 pp, medium confidence
BioMysteryBench (human-solvable)87.8%estimated ± 0.6 pp, medium confidence
FrontierCode 1.1 Extended59.4%estimated ± 1.6 pp, medium confidence
HealthBench (raw)51.2%estimated ± 6.0 pp, medium confidence
HealthBench (length-adjusted)53.8%estimated ± 0.9 pp, medium confidence
HealthBench Professional56.8%estimated ± 3.6 pp, high confidence
HealthBench Professional (raw)61.6%estimated ± 4.5 pp, high confidence
HLE-Verified53.5%estimated ± 2.0 pp, medium confidence
LABBench282.7%estimated ± 1.8 pp, low confidence
ScarfBench61.7%estimated ± 3.8 pp, low confidence
BlenderWorld (text-only)70.6%estimated ± 4.0 pp, medium confidence
Cookie-Bench81.5%estimated ± 2.5 pp, medium confidence
STEMBench2.6%estimated ± 0.2 pp, medium confidence
SimpleQA35.2%estimated ± 7.8 pp, low confidence
Agents' Last Exam30.9%estimated ± 9.7 pp, medium confidence
BabyVision w/ Python91.1%estimated ± 4.0 pp, low confidence
OpenHarmony Bench v1.058.9%estimated ± 2.4 pp, high confidence
SEC-Bench Pro33.0%estimated ± 10.7 pp, low confidence
MRCRv285.7%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.4%estimated ± 2.6 pp, low confidence
BioBasicsBench34.1%estimated ± 2.8 pp, medium confidence
LifeVal49.0%estimated ± 5.0 pp, low confidence
MMMU-Pro w/ Python83.1%estimated ± 2.0 pp, high confidence
ExploitBench Internal Port0.2%estimated ± 1.4 pp, low confidence
SWE-Atlas Codebase QnA49.8%estimated ± 4.8 pp, high confidence