benchgap
Anthropic · model

Claude Opus 4.8 benchmark scores

As of 2026-11-21, Claude Opus 4.8 (Anthropic) has measured scores on 70 benchmarks and estimated scores on 133 more.

BenchmarkScoreSource
AA-GPQA Diamond92.0%measured
AA-HLE48.7%measured
AA-SciCode54.4%measured
AIME2697.9%estimated ± 1.4 pp, medium confidence
Artificial Analysis Intelligence Index41.8%measured
CritPt20.9%measured
GDPval-AA47.7%measured
GPQA Diamond93.6%measured
HLE57.9%measured
IFBench61.5%estimated ± 6.6 pp, medium confidence
AA-LCR77.7%measured
LiveCodeBench v689.9%estimated ± 2.2 pp, low confidence
AA-Omniscience Accuracy48.8%measured
SciCode52.3%estimated ± 2.9 pp, low confidence
Terminal-Bench 2.174.6%measured
BrowseComp84.3%measured
HLE w/ tools56.9%estimated ± 5.3 pp, medium confidence
IFEval92.4%estimated ± 2.8 pp, high confidence
LongBench v263.8%estimated ± 5.5 pp, medium confidence
VITA-Bench46.0%estimated ± 6.7 pp, low confidence
τ²-bench results94.4%measured
AA Coding Index74.3%measured
AA-MMMU-Pro82.6%estimated ± 2.3 pp, high confidence
APEX-Agents40.2%estimated ± 9.4 pp, low confidence
APEX-Agents-AA36.5%estimated ± 2.9 pp, medium confidence
BioMysteryBench (human-difficult)48.3%estimated ± 6.8 pp, low confidence
DeepSearchQA93.1%measured
SWE-bench Verified88.6%measured
BFCL v476.5%estimated ± 2.7 pp, medium confidence
CyberGym78.3%estimated ± 5.3 pp, medium confidence
JobBench53.4%estimated ± 4.9 pp, medium confidence
SkillsBench56.2%estimated ± 4.1 pp, medium confidence
SWE-bench Pro69.2%measured
WideResearch81.4%estimated ± 2.1 pp, medium confidence
AA Agentic Index42.6%measured
MMLU-Pro85.8%estimated ± 2.5 pp, low confidence
AA-IFBench62.2%measured
FailureSensorIQ68.1%estimated ± 1.7 pp, high confidence
GPQA93.6%measured
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond92.4%measured
Vals LiveCodeBench87.8%measured
Vals MMLU-Pro89.6%measured
MMMLU90.9%estimated ± 0.7 pp, low confidence
MMMU90.5%estimated ± 3.7 pp, low confidence
Chess Puzzles52.0%measured
AA MMLU-Pro88.6%estimated ± 0.7 pp, medium confidence
Gert Labs73.0%measured
AA-AnalystAgent45.0%measured
AA EnterpriseOps-Gym44.0%measured
AA Harvey LAB v1.092.7%estimated ± 1.2 pp, medium confidence
AA Tau3 Banking98.8%estimated ± 2.0 pp, low confidence
AA Terminal-Bench 2.184.6%measured
ApprenticeBench20.0%estimated ± 3.1 pp, low confidence
ARC-AGI-192.5%measured
ARC-AGI-272.1%measured
BabaIsBench96.0%measured
CursorBench 3.158.4%measured
CursorBench 3.262.3%measured
EBR-bench28.6%measured
Terminal-Bench 3.021.1%measured
FrontierCode 1.1 Main46.5%measured
FrontierMath v2 (Tier 4)56.1%measured
FrontierMath v2 (Tiers 1-3)80.0%measured
FrontierSWE v226.1%estimated ± 8.0 pp, low confidence
Furniture Assembly42.5%measured
MaxxBench57.4%estimated ± 4.3 pp, low confidence
MazeBench0.0%measured
MirrorCode65.6%measured
Mystery Game Puzzles36.0%measured
NYT Connections91.1%measured
OfficeQA Pro66.2%measured
OSWorld-Verified83.4%measured
ParseBench-162.3%measured
Terminal-Bench 4.123.6%measured
Terminal-Bench Hard58.3%measured
Vals SWE-bench88.6%measured
Terminal-Bench 2.1 (Vals AI)71.9%measured
VulcanBench v382.7%estimated ± 5.1 pp, low confidence
CWE-Bench-AA46.8%estimated ± 4.5 pp, medium confidence
AA Cyber Index25.7%estimated ± 9.2 pp, low confidence
DeepsecBench-AA21.1%estimated ± 8.9 pp, low confidence
Harvey LAB-AA v1.17.4%estimated ± 4.0 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)58.6%estimated ± 4.3 pp, medium confidence
HealthBench Professional (raw, Artificial Analysis)66.0%estimated ± 0.8 pp, medium confidence
AA ITBench46.5%estimated ± 2.8 pp, medium confidence
Bug Hunt Bench27.3%estimated ± 6.3 pp, low confidence
Chartography (tools)86.0%estimated ± 4.3 pp, low confidence
CursorBench 4.038.5%estimated ± 2.0 pp, high confidence
CWE-bench v157.4%estimated ± 5.1 pp, medium confidence
DeepSWE66.2%estimated ± 4.3 pp, high confidence
DRACO86.9%estimated ± 0.7 pp, low confidence
FrontierCode 2.015.3%measured
FrontierPhysics3.5%estimated ± 4.4 pp, high confidence
HLE w/o tools49.8%measured
OfficeQA77.4%estimated ± 0.2 pp, medium confidence
OSWorld 2.020.6%measured
OSWorld 2.162.5%estimated ± 5.3 pp, low confidence
OSWorld 2.1 Offline52.9%estimated ± 1.9 pp, medium confidence
PostTrainBench v1.132.9%measured
ProgramBench87.9%estimated ± 4.8 pp, medium confidence
SpireClimb5.8%estimated ± 7.2 pp, medium confidence
StealthMark42.4%estimated ± 3.5 pp, high confidence
SWE Multilingual84.4%measured
SWE-bench Multimodal38.4%measured
Terminal-Bench 4.029.6%estimated ± 4.5 pp, high confidence
Terminal-Bench-Science 0.10.0%estimated ± 10.7 pp, low confidence
Toolathlon-Verified73.4%estimated ± 1.0 pp, medium confidence
VulcanBench Frontier v484.5%estimated ± 2.3 pp, medium confidence
AA AutomationBench52.8%estimated ± 12.8 pp, low confidence
GDP.pdf17.0%estimated ± 4.6 pp, high confidence
AA Terminal-Bench 4.023.2%estimated ± 6.2 pp, medium confidence
ARC-AGI-31.5%measured
ARC-AGI-3 Provider Adapter7.6%estimated ± 4.1 pp, high confidence
cursorBench4144.8%estimated ± 3.7 pp, medium confidence
CWE-bench55.8%estimated ± 4.6 pp, high confidence
MKA42.3%estimated ± 5.6 pp, low confidence
SkillsBench official leaderboard (BenchFlow)54.1%measured
BinaryAudit38.8%estimated ± 12.0 pp, low confidence
CompileBench100.0%estimated ± 10.0 pp, low confidence
OTelBench22.5%estimated ± 6.6 pp, medium confidence
Vibe Code Bench75.2%estimated ± 5.6 pp, low confidence
FACTS Multimodal50.9%estimated ± 4.0 pp, high confidence
Hermes Bench69.3%estimated ± 2.9 pp, high confidence
Hermes Index37.2%estimated ± 3.1 pp, high confidence
ParseBench-269.9%estimated ± 4.3 pp, high confidence
CharXiv89.9%measured
CharXiv w/o tools80.5%measured
ExploitBench41.0%estimated ± 1.9 pp, medium confidence
ExploitGym12.1%estimated ± 2.7 pp, high confidence
AI2D_TEST93.5%estimated ± 0.9 pp, low confidence
C-Eval98.1%estimated ± 1.0 pp, low confidence
Claw-Eval70.0%estimated ± 2.9 pp, medium confidence
CountBench98.5%estimated ± 3.7 pp, low confidence
DeepPlanning51.1%estimated ± 13.1 pp, low confidence
DynaMath88.0%estimated ± 0.8 pp, low confidence
ERQA63.6%estimated ± 2.9 pp, medium confidence
HMMT Feb 2026100.0%estimated ± 1.5 pp, low confidence
HMMT Nov 202595.8%estimated ± 1.5 pp, low confidence
INCLUDE87.6%measured
MathVision91.2%estimated ± 3.3 pp, high confidence
MCP Atlas82.2%measured
MLVU (M-Avg)88.9%estimated ± 1.5 pp, high confidence
MMAnswerBench94.2%estimated ± 2.4 pp, low confidence
MMLU-ProX88.7%estimated ± 1.3 pp, medium confidence
MMLU-Redux97.1%estimated ± 0.8 pp, low confidence
MMMU-Pro81.2%estimated ± 3.1 pp, high confidence
NL2Repo52.4%estimated ± 4.7 pp, high confidence
PostTrainBench v134.1%measured
QwenClawBench67.7%estimated ± 2.6 pp, medium confidence
RealWorldQA88.7%estimated ± 1.4 pp, low confidence
ScreenSpot Pro87.9%measured
SuperGPQA96.7%estimated ± 7.2 pp, low confidence
Terminal-Bench 2.075.9%estimated ± 5.4 pp, medium confidence
Toolathlon59.9%measured
V*96.3%estimated ± 6.5 pp, low confidence
Video-MME (with subtitle)90.7%estimated ± 2.9 pp, high confidence
VideoMMMU87.2%estimated ± 1.5 pp, medium confidence
We-Math84.6%estimated ± 7.9 pp, low confidence
AA LiveCodeBench96.9%estimated ± 2.6 pp, medium confidence
HealthBench Hard36.1%estimated ± 5.1 pp, low confidence
LiveCodeBench Pro90.4%estimated ± 6.0 pp, low confidence
MedXpertQA (MM)79.0%estimated ± 6.4 pp, low confidence
MedXpertQA (Text)66.3%estimated ± 4.5 pp, medium confidence
MMLU-Pro (Arcee)88.1%estimated ± 3.4 pp, low confidence
React Native Evals83.5%estimated ± 2.7 pp, low confidence
ResearchClawBench21.1%measured
SWE-Rebench61.8%estimated ± 3.6 pp, low confidence
PostTrainBench31.0%measured
DAYJOB2.0%measured
DAYJOB Healthcare0.8%measured
Finance Agent v253.9%measured
BioMysteryBench (human-solvable)89.2%estimated ± 0.6 pp, medium confidence
FrontierCode 1.1 Extended55.2%estimated ± 1.6 pp, medium confidence
HealthBench (raw)56.3%estimated ± 6.0 pp, medium confidence
HealthBench (length-adjusted)53.8%estimated ± 0.9 pp, medium confidence
HealthBench Professional57.0%estimated ± 3.6 pp, high confidence
HealthBench Professional (raw)66.5%estimated ± 4.5 pp, high confidence
HLE-Verified46.8%estimated ± 2.0 pp, medium confidence
LABBench276.3%estimated ± 1.7 pp, low confidence
ScarfBench0.8%estimated ± 5.0 pp, low confidence
BlenderWorld (text-only)72.4%estimated ± 2.8 pp, medium confidence
Cookie-Bench82.0%estimated ± 1.3 pp, medium confidence
STEMBench2.6%estimated ± 0.2 pp, medium confidence
SimpleQA34.6%estimated ± 7.8 pp, low confidence
LiveCodeBench84.2%estimated ± 14.6 pp, low confidence
Agents' Last Exam35.4%estimated ± 11.0 pp, low confidence
BabyVision w/ Python87.2%estimated ± 3.7 pp, low confidence
OpenHarmony Bench v1.058.9%estimated ± 2.4 pp, high confidence
SEC-Bench Pro50.1%estimated ± 5.8 pp, medium confidence
BabyVision50.5%estimated ± 7.4 pp, low confidence
MMVU80.6%estimated ± 3.6 pp, medium confidence
PerceptionBench55.3%estimated ± 3.8 pp, low confidence
ZeroBench37.0%estimated ± 6.7 pp, low confidence
MRCRv285.4%estimated ± 11.3 pp, low confidence
LVBench70.6%estimated ± 8.2 pp, low confidence
MMLongBench-Doc61.3%estimated ± 2.6 pp, low confidence
BioBasicsBench37.2%estimated ± 2.2 pp, medium confidence
LifeVal75.6%estimated ± 1.6 pp, medium confidence
MMMU-Pro w/ Python80.3%estimated ± 2.7 pp, medium confidence
ExploitBench Internal Port2.5%estimated ± 1.4 pp, medium confidence
MathVision w/ Python95.7%estimated ± 1.0 pp, low confidence
PinchBench90.4%estimated ± 3.6 pp, low confidence