benchgap
Anthropic · model

Claude Sonnet 4.6 benchmark scores

As of 2026-11-22, Claude Sonnet 4.6 (Anthropic) has measured scores on 40 benchmarks and estimated scores on 130 more.

BenchmarkScoreSource
AA-GPQA Diamond79.9%measured
AA-HLE13.3%measured
AA-SciCode52.2%estimated ± 1.7 pp, medium confidence
Artificial Analysis Intelligence Index24.7%measured
CritPt0.9%measured
GDPval-AA34.5%estimated ± 5.0 pp, high confidence
GPQA Diamond89.9%estimated ± 0.1 pp, high confidence
HLE49.0%measured
IFBench54.3%estimated ± 6.6 pp, low confidence
AA-LCR68.3%measured
LiveCodeBench v687.4%estimated ± 2.2 pp, medium confidence
AA-Omniscience Accuracy38.6%measured
SciCode40.9%estimated ± 2.9 pp, low confidence
Terminal-Bench 2.174.8%estimated ± 6.7 pp, medium confidence
BrowseComp82.9%estimated ± 1.9 pp, high confidence
HLE w/ tools50.0%estimated ± 4.9 pp, medium confidence
IFEval89.2%estimated ± 2.8 pp, high confidence
LongBench v257.6%estimated ± 5.5 pp, medium confidence
VITA-Bench44.9%estimated ± 10.7 pp, low confidence
τ²-bench results79.5%measured
AA Coding Index66.0%estimated ± 3.6 pp, medium confidence
AA-MMMU-Pro70.6%measured
APEX-Agents-AA32.4%estimated ± 2.9 pp, medium confidence
BioMysteryBench (human-difficult)15.7%estimated ± 6.8 pp, low confidence
DeepSearchQA73.4%estimated ± 7.5 pp, low confidence
SWE-bench Verified79.6%measured
AutomationBench18.0%estimated ± 10.4 pp, low confidence
BFCL v461.1%estimated ± 8.5 pp, low confidence
CyberGym65.2%measured
JobBench36.9%measured
SWE-bench Pro57.0%estimated ± 2.9 pp, medium confidence
WideResearch78.4%estimated ± 1.2 pp, medium confidence
AA Agentic Index28.7%estimated ± 5.8 pp, medium confidence
MMLU-Pro79.2%measured
AA-IFBench41.2%measured
BBH95.4%estimated ± 2.8 pp, medium confidence
FailureSensorIQ64.7%estimated ± 1.7 pp, high confidence
GPQA89.9%measured
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond85.6%measured
Vals LiveCodeBench82.1%measured
Vals MMLU-Pro87.3%measured
MMMLU86.3%estimated ± 0.7 pp, medium confidence
MMMU80.5%estimated ± 2.7 pp, high confidence
Chess Puzzles23.0%measured
AA MMLU-Pro87.1%estimated ± 0.7 pp, high confidence
Gert Labs47.3%measured
AA-AnalystAgent44.8%estimated ± 5.9 pp, low confidence
AA EnterpriseOps-Gym45.0%estimated ± 1.2 pp, medium confidence
AA Harvey LAB v1.088.9%estimated ± 1.3 pp, medium confidence
AA Terminal-Bench 2.180.3%estimated ± 2.5 pp, medium confidence
ApprenticeBench2.0%measured
ARC-AGI-186.0%measured
ARC-AGI-258.3%measured
CursorBench 3.148.8%measured
CursorBench 3.251.0%estimated ± 5.0 pp, low confidence
EBR-bench9.4%estimated ± 8.1 pp, low confidence
Terminal-Bench 3.04.4%estimated ± 6.9 pp, low confidence
FrontierCode 1.1 Main24.3%measured
FrontierSWE v29.0%estimated ± 9.5 pp, low confidence
Furniture Assembly32.9%estimated ± 9.3 pp, medium confidence
MaxxBench47.2%estimated ± 8.8 pp, medium confidence
MazeBench0.0%measured
Mystery Game Puzzles16.0%measured
NYT Connections44.8%measured
OfficeQA Pro50.3%estimated ± 6.8 pp, low confidence
OSWorld-Verified72.1%measured
ParseBench-154.4%estimated ± 6.2 pp, medium confidence
Terminal-Bench 4.110.5%estimated ± 14.9 pp, low confidence
Terminal-Bench Hard46.8%estimated ± 3.9 pp, high confidence
Vals SWE-bench77.4%measured
Terminal-Bench 2.1 (Vals AI)57.3%measured
VulcanBench v381.9%estimated ± 5.6 pp, low confidence
CWE-Bench-AA43.3%estimated ± 4.5 pp, medium confidence
AA Cyber Index39.4%estimated ± 12.4 pp, low confidence
DeepsecBench-AA16.2%estimated ± 8.2 pp, medium confidence
HealthBench Professional (length-adjusted, Artificial Analysis)53.3%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)59.1%estimated ± 1.1 pp, low confidence
AA ITBench31.9%estimated ± 4.4 pp, low confidence
Chartography (tools)77.4%estimated ± 4.1 pp, low confidence
CursorBench 4.030.9%estimated ± 3.4 pp, medium confidence
CWE-bench v157.4%estimated ± 5.1 pp, medium confidence
DeepSWE56.3%estimated ± 7.2 pp, low confidence
DRACO86.9%estimated ± 0.7 pp, low confidence
FrontierPhysics2.5%estimated ± 5.0 pp, low confidence
HLE w/o tools39.2%estimated ± 4.9 pp, high confidence
OSWorld 2.08.3%measured
OSWorld 2.157.8%estimated ± 3.6 pp, high confidence
OSWorld 2.1 Offline56.6%estimated ± 10.1 pp, low confidence
PostTrainBench v1.121.9%estimated ± 3.7 pp, medium confidence
SpireClimb5.8%estimated ± 7.2 pp, low confidence
StealthMark36.8%estimated ± 9.2 pp, medium confidence
Toolathlon-Verified62.8%estimated ± 1.0 pp, low confidence
AA AutomationBench52.7%estimated ± 9.9 pp, low confidence
GDP.pdf16.4%estimated ± 4.1 pp, medium confidence
AA Terminal-Bench 4.03.7%estimated ± 13.7 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-bench45.3%estimated ± 6.2 pp, medium confidence
MKA35.0%estimated ± 4.6 pp, high confidence
SkillsBench official leaderboard (BenchFlow)47.2%measured
BinaryAudit31.4%estimated ± 12.0 pp, low confidence
CompileBench75.5%estimated ± 10.0 pp, low confidence
OTelBench18.5%estimated ± 4.1 pp, medium confidence
Vibe Code Bench51.5%measured
FACTS Multimodal26.6%estimated ± 4.0 pp, medium confidence
Hermes Index37.2%estimated ± 3.1 pp, high confidence
ParseBench-258.7%estimated ± 6.2 pp, medium confidence
CharXiv77.4%measured
CharXiv w/o tools77.8%estimated ± 2.4 pp, medium confidence
ExploitBench30.5%estimated ± 7.2 pp, low confidence
ExploitGym11.9%estimated ± 6.3 pp, medium confidence
AI2D_TEST90.8%estimated ± 0.9 pp, medium confidence
C-Eval93.0%estimated ± 1.0 pp, medium confidence
Claw-Eval67.8%measured
CountBench91.9%estimated ± 2.9 pp, low confidence
DynaMath84.7%estimated ± 0.8 pp, medium confidence
ERQA45.0%estimated ± 3.1 pp, medium confidence
MathVision78.9%estimated ± 3.1 pp, high confidence
MCP Atlas72.3%estimated ± 5.8 pp, low confidence
MLVU (M-Avg)84.8%estimated ± 1.5 pp, high confidence
MMLU-Redux89.9%estimated ± 0.6 pp, high confidence
MMMU-Pro72.0%estimated ± 1.5 pp, high confidence
NL2Repo43.8%estimated ± 4.7 pp, high confidence
PostTrainBench v120.0%estimated ± 3.1 pp, medium confidence
RealWorldQA82.3%estimated ± 1.5 pp, high confidence
ScreenSpot Pro66.9%estimated ± 6.8 pp, medium confidence
SuperGPQA95.0%measured
Terminal-Bench 2.059.1%measured
Toolathlon51.9%estimated ± 1.7 pp, low confidence
V*58.0%estimated ± 7.5 pp, low confidence
Video-MME (w/o subtitle)80.1%estimated ± 1.7 pp, medium confidence
Video-MME (with subtitle)85.3%estimated ± 2.9 pp, high confidence
VideoMMMU84.9%estimated ± 1.5 pp, medium confidence
We-Math84.6%estimated ± 7.9 pp, low confidence
AA LiveCodeBench90.2%estimated ± 2.6 pp, medium confidence
HealthBench Hard1.7%estimated ± 5.1 pp, low confidence
MedXpertQA (MM)74.0%estimated ± 6.4 pp, low confidence
MMLU-Pro (Arcee)80.5%estimated ± 4.0 pp, medium confidence
React Native Evals80.6%measured
ResearchClawBench18.3%estimated ± 1.3 pp, high confidence
SWE-Rebench60.7%measured
PostTrainBench19.4%estimated ± 3.7 pp, medium confidence
DAYJOB1.3%estimated ± 3.2 pp, high confidence
DAYJOB Healthcare0.7%estimated ± 4.2 pp, high confidence
Finance Agent v255.6%estimated ± 2.4 pp, medium confidence
BioMysteryBench (human-solvable)81.7%estimated ± 0.6 pp, low confidence
HealthBench (raw)48.1%estimated ± 7.6 pp, low confidence
HealthBench (length-adjusted)53.5%estimated ± 2.5 pp, low confidence
HealthBench Professional50.9%estimated ± 4.2 pp, low confidence
HLE-Verified1.2%estimated ± 4.6 pp, low confidence
LABBench281.6%estimated ± 1.8 pp, low confidence
ScarfBench48.3%estimated ± 12.3 pp, low confidence
BlenderWorld (text-only)70.0%estimated ± 2.8 pp, medium confidence
Cookie-Bench76.9%estimated ± 1.3 pp, medium confidence
STEMBench2.5%estimated ± 0.2 pp, medium confidence
SimpleQA43.7%estimated ± 4.3 pp, low confidence
LiveCodeBench79.0%estimated ± 14.6 pp, low confidence
Agents' Last Exam22.5%estimated ± 3.3 pp, low confidence
BabyVision w/ Python84.7%estimated ± 4.0 pp, low confidence
OpenHarmony Bench v1.053.5%estimated ± 3.2 pp, high confidence
SEC-Bench Pro39.8%estimated ± 9.3 pp, medium confidence
BabyVision26.4%estimated ± 11.5 pp, low confidence
MMVU75.0%estimated ± 3.6 pp, medium confidence
MRCRv265.9%estimated ± 11.3 pp, low confidence
MMLongBench-Doc60.8%estimated ± 2.6 pp, low confidence
BioBasicsBench21.5%estimated ± 2.2 pp, low confidence
LifeVal75.2%estimated ± 3.8 pp, medium confidence
MMMU-Pro w/ Python74.3%estimated ± 2.0 pp, high confidence
ExploitBench Internal Port0.0%estimated ± 1.4 pp, low confidence