benchgap
Anthropic · model

Claude Opus 4.5 benchmark scores

As of 2026-11-22, Claude Opus 4.5 (Anthropic) has measured scores on 80 benchmarks and estimated scores on 115 more.

BenchmarkScoreSource
AA-GPQA Diamond81.0%measured
AA-HLE13.2%measured
AA-SciCode47.6%estimated ± 3.1 pp, high confidence
AIME2695.1%measured
Artificial Analysis Intelligence Index23.7%measured
CritPt0.3%measured
GDPval-AA27.4%estimated ± 5.0 pp, high confidence
GPQA Diamond87.0%estimated ± 0.1 pp, medium confidence
HLE30.8%measured
IFBench58.0%measured
AA-LCR70.7%measured
LiveCodeBench v684.8%measured
AA-Omniscience Accuracy40.9%measured
SciCode48.0%estimated ± 3.6 pp, high confidence
BrowseComp73.4%estimated ± 4.3 pp, high confidence
HLE w/ tools50.0%estimated ± 4.9 pp, medium confidence
IFEval90.9%measured
LongBench v264.4%measured
VITA-Bench23.3%measured
τ²-bench results86.3%measured
AA Coding Index61.7%estimated ± 4.3 pp, high confidence
AA-MMMU-Pro71.2%measured
APEX-Agents0.0%estimated ± 12.6 pp, low confidence
APEX-Agents-AA30.9%estimated ± 3.6 pp, medium confidence
BioMysteryBench (human-difficult)6.5%estimated ± 6.8 pp, low confidence
DeepSearchQA81.1%estimated ± 7.3 pp, low confidence
SWE-bench Verified80.9%measured
AutomationBench48.6%estimated ± 9.1 pp, low confidence
BFCL v467.5%estimated ± 4.7 pp, low confidence
CyberGym50.6%measured
JobBench32.3%measured
SkillsBench51.6%estimated ± 4.9 pp, medium confidence
SWE-bench Pro57.1%measured
τ³-bench results70.2%measured
WideResearch76.4%measured
AA Agentic Index18.8%estimated ± 5.8 pp, low confidence
MMLU-Pro89.5%measured
AA-IFBench43.0%measured
BBH100.0%estimated ± 2.8 pp, medium confidence
FailureSensorIQ66.7%measured
GPQA87.0%measured
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond79.5%measured
Vals LiveCodeBench75.0%measured
Vals MMLU-Pro85.6%measured
MATH-50097.7%estimated ± 3.1 pp, medium confidence
MMMLU85.4%estimated ± 0.7 pp, medium confidence
MMMU80.7%measured
Chess Puzzles14.0%measured
AIME 202596.4%estimated ± 1.8 pp, medium confidence
AA MMLU-Pro87.5%estimated ± 0.7 pp, medium confidence
Gert Labs46.0%measured
AA EnterpriseOps-Gym44.5%estimated ± 1.2 pp, low confidence
AA Tau3 Banking39.6%estimated ± 2.0 pp, low confidence
ApprenticeBench4.1%estimated ± 4.0 pp, medium confidence
ARC-AGI-166.5%estimated ± 6.5 pp, medium confidence
ARC-AGI-234.1%estimated ± 10.9 pp, low confidence
BabaIsBench91.8%estimated ± 1.2 pp, high confidence
CursorBench 3.152.7%estimated ± 5.6 pp, low confidence
CursorBench 3.253.9%estimated ± 3.6 pp, high confidence
EBR-bench14.3%measured
FrontierCode 1.1 Main29.1%estimated ± 4.2 pp, medium confidence
FrontierMath v2 (Tier 4)4.9%measured
FrontierMath v2 (Tiers 1-3)34.4%measured
FrontierSWE v210.6%estimated ± 8.3 pp, medium confidence
Furniture Assembly28.3%measured
MaxxBench42.0%estimated ± 4.3 pp, low confidence
MazeBench0.4%estimated ± 2.7 pp, high confidence
Mystery Game Puzzles22.0%measured
NYT Connections49.4%measured
OfficeQA Pro50.3%estimated ± 6.8 pp, low confidence
OSWorld-Verified66.3%measured
ParseBench-139.9%estimated ± 5.2 pp, low confidence
Terminal-Bench Hard44.4%estimated ± 4.0 pp, high confidence
Vals SWE-bench74.6%estimated ± 2.8 pp, medium confidence
Terminal-Bench 2.1 (Vals AI)52.3%estimated ± 6.4 pp, medium confidence
VulcanBench v379.1%estimated ± 5.6 pp, low confidence
CWE-Bench-AA50.1%estimated ± 8.1 pp, medium confidence
AA Cyber Index38.9%estimated ± 12.4 pp, low confidence
DeepsecBench-AA15.5%estimated ± 8.2 pp, medium confidence
HealthBench Professional (length-adjusted, Artificial Analysis)53.0%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)58.4%estimated ± 1.1 pp, low confidence
AA ITBench27.1%estimated ± 4.4 pp, medium confidence
Bug Hunt Bench12.1%estimated ± 7.2 pp, low confidence
Chartography (tools)77.8%estimated ± 4.1 pp, low confidence
CursorBench 4.033.4%estimated ± 3.3 pp, medium confidence
CWE-bench v148.3%estimated ± 5.7 pp, medium confidence
DeepSWE40.8%estimated ± 8.7 pp, medium confidence
FrontierCode 2.020.3%estimated ± 6.7 pp, low confidence
FrontierPhysics0.2%estimated ± 4.4 pp, high confidence
HLE w/o tools32.3%estimated ± 4.9 pp, high confidence
OSWorld 2.00.1%estimated ± 3.6 pp, medium confidence
OSWorld 2.154.9%estimated ± 3.6 pp, medium confidence
OSWorld 2.1 Offline55.4%estimated ± 10.1 pp, low confidence
PostTrainBench v1.112.9%estimated ± 1.3 pp, low confidence
SpireClimb5.8%estimated ± 7.2 pp, low confidence
StealthMark45.3%estimated ± 6.5 pp, medium confidence
SWE Multilingual77.5%measured
Toolathlon-Verified71.3%estimated ± 3.0 pp, high confidence
VulcanBench Frontier v478.0%estimated ± 8.3 pp, low confidence
AA AutomationBench52.0%estimated ± 9.9 pp, low confidence
GDP.pdf16.4%estimated ± 4.1 pp, low confidence
ARC-AGI-30.0%estimated ± 8.3 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
CWE-bench44.4%estimated ± 6.2 pp, medium confidence
MKA34.6%estimated ± 4.6 pp, high confidence
SkillsBench official leaderboard (BenchFlow)49.0%measured
BinaryAudit36.8%measured
CompileBench88.9%estimated ± 10.0 pp, low confidence
OTelBench29.0%measured
Vibe Code Bench55.5%estimated ± 8.4 pp, medium confidence
FACTS Multimodal29.0%estimated ± 4.0 pp, medium confidence
Hermes Index30.1%estimated ± 5.7 pp, low confidence
ParseBench-259.0%estimated ± 5.0 pp, medium confidence
CharXiv68.5%measured
CharXiv w/o tools76.1%estimated ± 1.2 pp, high confidence
ExploitBench32.4%estimated ± 6.5 pp, low confidence
ExploitGym1.9%estimated ± 8.3 pp, low confidence
AI2D_TEST87.7%measured
C-Eval92.2%measured
Claw-Eval59.6%measured
CountBench90.6%measured
DeepPlanning26.4%measured
DynaMath79.7%measured
ERQA46.8%measured
HMMT Feb 202592.9%measured
HMMT Feb 202685.3%measured
HMMT Nov 202593.3%measured
INCLUDE86.2%measured
MathVision74.3%measured
MAXIFE79.2%measured
MCP Atlas42.3%measured
MCP-Tasks71.8%measured
MLVU (M-Avg)81.7%measured
MMAnswerBench84.0%measured
MMLU-ProX85.7%measured
MMLU-Redux96.6%measured
MMMU-Pro70.6%measured
NL2Repo43.2%measured
NOVA-6356.7%measured
OmniDocBench 1.587.7%measured
PolyMath79.0%measured
PostTrainBench v117.1%measured
QwenClawBench52.3%measured
RealWorldQA77.0%measured
ScreenSpot Pro45.7%measured
SimpleVQA65.7%measured
SuperGPQA70.6%measured
Terminal-Bench 2.059.3%measured
Toolathlon43.5%measured
V*67.0%measured
Video-MME (w/o subtitle)81.4%measured
Video-MME (with subtitle)77.6%measured
VideoMMMU84.4%measured
We-Math70.0%measured
HealthBench Hard0.0%estimated ± 5.1 pp, low confidence
LiveCodeBench Pro81.4%estimated ± 6.0 pp, low confidence
MedXpertQA (MM)52.4%estimated ± 2.4 pp, high confidence
MMLU-Pro (Arcee)78.3%estimated ± 4.0 pp, low confidence
React Native Evals80.1%estimated ± 2.7 pp, medium confidence
ResearchClawBench16.3%estimated ± 1.3 pp, high confidence
SWE-Rebench61.8%estimated ± 3.6 pp, medium confidence
PostTrainBench14.8%estimated ± 1.2 pp, low confidence
DAYJOB1.4%estimated ± 2.7 pp, high confidence
DAYJOB Healthcare0.0%estimated ± 4.0 pp, high confidence
Finance Agent v255.5%estimated ± 2.4 pp, medium confidence
BioMysteryBench (human-solvable)81.7%estimated ± 0.6 pp, low confidence
FrontierCode 1.1 Extended55.0%estimated ± 3.0 pp, low confidence
HealthBench (length-adjusted)54.1%estimated ± 3.7 pp, medium confidence
HealthBench Professional56.8%estimated ± 4.2 pp, low confidence
HLE-Verified35.7%estimated ± 7.1 pp, low confidence
LABBench280.7%estimated ± 1.8 pp, low confidence
BlenderWorld (text-only)70.2%estimated ± 2.8 pp, medium confidence
Cookie-Bench77.3%estimated ± 1.3 pp, medium confidence
STEMBench2.5%estimated ± 0.2 pp, medium confidence
SimpleQA46.8%estimated ± 4.3 pp, low confidence
LiveCodeBench79.8%estimated ± 14.6 pp, low confidence
Agents' Last Exam30.7%estimated ± 9.7 pp, medium confidence
BabyVision w/ Python88.8%estimated ± 3.8 pp, low confidence
OpenHarmony Bench v1.051.2%estimated ± 2.7 pp, high confidence
SEC-Bench Pro29.6%estimated ± 7.5 pp, low confidence
IMOAnswerBench80.6%estimated ± 5.6 pp, medium confidence
BabyVision22.7%estimated ± 3.7 pp, medium confidence
MMVU72.5%estimated ± 2.7 pp, medium confidence
PerceptionBench55.6%estimated ± 4.9 pp, low confidence
MRCRv276.6%estimated ± 11.3 pp, low confidence
CC-OCR76.1%estimated ± 3.7 pp, medium confidence
MMLongBench-Doc61.0%estimated ± 2.6 pp, low confidence
RefCOCO (avg)88.4%estimated ± 3.1 pp, medium confidence
OCRBench V270.6%estimated ± 2.1 pp, medium confidence
BioBasicsBench34.1%estimated ± 3.8 pp, medium confidence
LifeVal45.6%estimated ± 5.0 pp, low confidence
MMMU-Pro w/ Python74.8%estimated ± 2.0 pp, high confidence
ExploitBench Internal Port0.0%estimated ± 1.4 pp, low confidence
PinchBench94.5%estimated ± 5.5 pp, low confidence