benchgap
Anthropic · model

Claude Haiku 5.5 benchmark scores

As of 2026-11-22, Claude Haiku 5.5 (Anthropic) has measured scores on 49 benchmarks and estimated scores on 113 more.

BenchmarkScoreSource
AA-GPQA Diamond90.9%estimated ± 2.7 pp, high confidence
AA-HLE44.4%measured
AA-SciCode55.0%measured
AIME2699.2%estimated ± 1.2 pp, low confidence
Artificial Analysis Intelligence Index36.5%measured
CritPt18.9%measured
GDPval-AA56.0%measured
GPQA Diamond90.6%estimated ± 1.4 pp, high confidence
HLE54.3%estimated ± 2.2 pp, high confidence
AA-LCR82.7%measured
LiveCodeBench v695.5%estimated ± 4.9 pp, medium confidence
AA-Omniscience Accuracy34.7%measured
SciCode54.7%estimated ± 3.1 pp, high confidence
Terminal-Bench 2.189.1%estimated ± 5.4 pp, low confidence
BrowseComp82.2%estimated ± 7.1 pp, medium confidence
HLE w/ tools57.4%measured
LongBench v268.1%estimated ± 5.5 pp, medium confidence
τ²-bench results98.1%estimated ± 3.1 pp, high confidence
AA Coding Index79.7%estimated ± 1.2 pp, high confidence
AA-MMMU-Pro79.6%estimated ± 5.4 pp, medium confidence
APEX-Agents50.4%estimated ± 9.4 pp, low confidence
APEX-Agents-AA37.7%estimated ± 9.3 pp, medium confidence
BioMysteryBench (human-difficult)47.8%estimated ± 6.8 pp, low confidence
DeepSearchQA89.6%estimated ± 9.6 pp, medium confidence
SWE-bench Verified90.4%estimated ± 2.2 pp, medium confidence
BFCL v477.1%estimated ± 2.7 pp, medium confidence
CyberGym88.2%estimated ± 6.0 pp, medium confidence
JobBench60.8%estimated ± 8.7 pp, medium confidence
SWE-bench Pro75.7%estimated ± 4.2 pp, high confidence
WideResearch83.0%estimated ± 3.3 pp, medium confidence
AA Agentic Index52.1%estimated ± 3.1 pp, high confidence
MMLU-Pro85.8%estimated ± 2.5 pp, low confidence
FailureSensorIQ66.3%estimated ± 1.7 pp, high confidence
GPQA93.1%estimated ± 0.9 pp, high confidence
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond93.0%estimated ± 1.4 pp, high confidence
Vals LiveCodeBench87.1%estimated ± 1.9 pp, high confidence
Vals MMLU-Pro89.0%estimated ± 1.6 pp, medium confidence
MMMLU90.7%estimated ± 0.7 pp, low confidence
Chess Puzzles55.0%measured
AA MMLU-Pro86.3%estimated ± 0.8 pp, high confidence
Gert Labs58.6%measured
AA-AnalystAgent53.8%measured
AA Harvey LAB v1.089.9%measured
AA Tau3 Banking45.5%estimated ± 4.1 pp, high confidence
ApprenticeBench13.8%estimated ± 10.4 pp, low confidence
ARC-AGI-189.0%measured
ARC-AGI-261.3%measured
BabaIsBench99.3%estimated ± 1.2 pp, high confidence
CursorBench 3.156.3%estimated ± 3.4 pp, medium confidence
CursorBench 3.263.4%estimated ± 1.6 pp, medium confidence
EBR-bench43.8%measured
Terminal-Bench 3.033.4%estimated ± 3.2 pp, high confidence
FrontierCode 1.1 Main46.4%measured
FrontierMath v2 (Tier 4)46.3%measured
FrontierMath v2 (Tiers 1-3)75.1%measured
FrontierSWE v243.8%measured
Furniture Assembly47.5%measured
MaxxBench65.7%estimated ± 4.3 pp, medium confidence
MazeBench1.0%measured
MirrorCode78.9%measured
Mystery Game Puzzles30.0%measured
NYT Connections65.7%measured
OfficeQA Pro51.9%estimated ± 7.0 pp, low confidence
OSWorld-Verified85.1%estimated ± 8.4 pp, low confidence
ParseBench-156.1%measured
Terminal-Bench 4.140.0%measured
Terminal-Bench Hard63.6%estimated ± 2.8 pp, medium confidence
Vals SWE-bench86.6%estimated ± 2.8 pp, high confidence
Terminal-Bench 2.1 (Vals AI)72.8%estimated ± 6.9 pp, low confidence
VulcanBench v378.2%estimated ± 3.9 pp, medium confidence
CWE-Bench-AA63.3%estimated ± 3.8 pp, high confidence
AA Cyber Index41.4%estimated ± 12.4 pp, low confidence
DeepsecBench-AA0.0%estimated ± 7.5 pp, low confidence
Harvey LAB-AA v1.12.3%estimated ± 3.5 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)58.4%measured
HealthBench Professional (raw, Artificial Analysis)64.8%measured
AA ITBench46.6%estimated ± 4.6 pp, medium confidence
Bug Hunt Bench21.5%measured
Chartography (no tools)46.4%measured
Chartography (tools)86.3%estimated ± 2.7 pp, medium confidence
CursorBench 4.048.4%measured
CWE-bench v157.4%estimated ± 5.1 pp, medium confidence
DeepSWE72.2%estimated ± 4.1 pp, medium confidence
DRACO86.9%estimated ± 0.7 pp, low confidence
FrontierCode 2.028.9%measured
FrontierPhysics4.2%measured
HLE w/o tools45.9%measured
OSWorld 2.172.3%estimated ± 3.6 pp, high confidence
OSWorld 2.1 Offline72.4%measured
PostTrainBench v1.135.7%estimated ± 1.7 pp, medium confidence
SpireClimb5.8%estimated ± 7.2 pp, low confidence
StealthMark49.5%measured
SWE-bench Multimodal2.3%estimated ± 5.6 pp, low confidence
Terminal-Bench 4.039.2%measured
Terminal-Bench-Science 0.125.7%estimated ± 10.7 pp, low confidence
Toolathlon-Verified77.7%estimated ± 1.8 pp, medium confidence
VulcanBench Frontier v492.2%measured
AA AutomationBench35.4%measured
GDP.pdf20.8%measured
AA Terminal-Bench 4.032.8%measured
ARC-AGI-30.5%measured
ARC-AGI-3 Provider Adapter2.8%measured
cursorBench4148.4%measured
CWE-bench46.1%estimated ± 4.3 pp, medium confidence
MKA41.0%measured
SkillsBench official leaderboard (BenchFlow)55.3%estimated ± 6.1 pp, medium confidence
BinaryAudit35.8%estimated ± 12.1 pp, low confidence
CompileBench100.0%estimated ± 10.0 pp, low confidence
OTelBench32.1%estimated ± 4.1 pp, low confidence
Vibe Code Bench75.2%estimated ± 5.6 pp, low confidence
FACTS Multimodal38.9%measured
Hermes Bench71.0%measured
Hermes Index37.0%measured
ParseBench-271.2%measured
CharXiv82.3%estimated ± 5.6 pp, low confidence
CharXiv w/o tools75.1%estimated ± 1.8 pp, low confidence
ExploitBench33.2%estimated ± 6.5 pp, medium confidence
ExploitGym15.8%estimated ± 2.7 pp, high confidence
C-Eval96.5%estimated ± 1.0 pp, low confidence
HMMT Feb 2026100.0%estimated ± 1.5 pp, low confidence
HMMT Nov 202595.7%estimated ± 1.5 pp, low confidence
MCP Atlas82.2%estimated ± 10.7 pp, low confidence
MMAnswerBench93.0%estimated ± 2.4 pp, low confidence
MMLU-Redux95.3%estimated ± 0.8 pp, medium confidence
MMMU-Pro78.9%estimated ± 5.3 pp, low confidence
NL2Repo56.1%estimated ± 6.5 pp, low confidence
ScreenSpot Pro78.6%estimated ± 10.1 pp, low confidence
SuperGPQA73.1%estimated ± 9.3 pp, medium confidence
Toolathlon55.9%estimated ± 6.3 pp, low confidence
HealthBench Hard32.3%estimated ± 8.6 pp, low confidence
MedXpertQA (Text)63.2%estimated ± 8.9 pp, low confidence
React Native Evals78.8%estimated ± 4.7 pp, medium confidence
ResearchClawBench21.8%estimated ± 2.2 pp, low confidence
PostTrainBench33.1%estimated ± 2.4 pp, high confidence
DAYJOB1.4%estimated ± 2.7 pp, high confidence
DAYJOB Healthcare3.1%estimated ± 4.0 pp, high confidence
Finance Agent v256.9%estimated ± 2.4 pp, medium confidence
BioMysteryBench (human-solvable)88.9%estimated ± 0.6 pp, medium confidence
FrontierCode 1.1 Extended61.4%estimated ± 3.0 pp, medium confidence
HealthBench (raw)52.7%estimated ± 6.0 pp, medium confidence
HealthBench (length-adjusted)53.8%estimated ± 0.9 pp, medium confidence
HealthBench Professional59.4%estimated ± 2.1 pp, medium confidence
HealthBench Professional (raw)59.9%estimated ± 2.3 pp, medium confidence
HLE-Verified54.0%estimated ± 6.8 pp, low confidence
LABBench281.7%estimated ± 2.4 pp, medium confidence
ScarfBench68.7%estimated ± 4.5 pp, low confidence
BlenderWorld (text-only)70.9%estimated ± 2.8 pp, medium confidence
Cookie-Bench79.2%estimated ± 1.3 pp, medium confidence
STEMBench2.5%estimated ± 0.2 pp, medium confidence
SimpleQA38.3%estimated ± 4.3 pp, low confidence
Agents' Last Exam34.2%estimated ± 9.7 pp, medium confidence
OpenHarmony Bench v1.058.9%estimated ± 2.4 pp, high confidence
OpenHarmony Bench v1.263.8%estimated ± 11.0 pp, low confidence
SEC-Bench Pro35.8%estimated ± 5.6 pp, medium confidence
MRCRv285.9%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.5%estimated ± 2.6 pp, low confidence
SWE-Marathon43.7%estimated ± 2.1 pp, medium confidence
BioBasicsBench34.5%estimated ± 2.8 pp, medium confidence
LifeVal82.2%estimated ± 2.3 pp, medium confidence
MMMU-Pro w/ Python77.6%estimated ± 2.7 pp, medium confidence
ExploitBench Internal Port3.8%estimated ± 1.4 pp, medium confidence