benchgap
Anthropic · model

Claude Fable 5 benchmark scores

As of 2026-11-22, Claude Fable 5 (Anthropic) has measured scores on 51 benchmarks and estimated scores on 121 more.

BenchmarkScoreSource
AA-GPQA Diamond92.6%measured
AA-HLE55.5%measured
AA-SciCode61.0%measured
AIME2698.0%estimated ± 1.4 pp, medium confidence
Artificial Analysis Intelligence Index49.6%measured
CritPt28.6%measured
GDPval-AA55.5%measured
GPQA Diamond96.4%estimated ± 1.4 pp, high confidence
HLE61.4%estimated ± 6.1 pp, medium confidence
IFBench62.4%estimated ± 6.6 pp, medium confidence
AA-LCR82.3%measured
LiveCodeBench v690.8%estimated ± 2.2 pp, low confidence
AA-Omniscience Accuracy65.4%measured
SciCode56.4%estimated ± 2.9 pp, low confidence
Terminal-Bench 2.184.3%measured
BrowseComp91.2%estimated ± 2.9 pp, high confidence
HLE w/ tools60.6%estimated ± 5.3 pp, medium confidence
IFEval92.4%estimated ± 2.8 pp, high confidence
LongBench v267.7%estimated ± 5.5 pp, medium confidence
τ²-bench results98.5%measured
AA Coding Index76.5%measured
AA-MMMU-Pro78.6%estimated ± 2.3 pp, high confidence
APEX-Agents49.7%estimated ± 9.4 pp, low confidence
APEX-Agents-AA46.0%estimated ± 7.8 pp, low confidence
BioMysteryBench (human-difficult)49.5%estimated ± 6.8 pp, low confidence
DeepSearchQA93.5%estimated ± 4.5 pp, high confidence
SWE-bench Verified95.0%measured
AutomationBench24.5%estimated ± 7.1 pp, low confidence
BFCL v477.0%estimated ± 2.7 pp, medium confidence
CyberGym79.9%estimated ± 3.1 pp, low confidence
JobBench60.2%estimated ± 8.7 pp, medium confidence
SWE-bench Pro80.0%measured
WideResearch81.4%estimated ± 2.1 pp, medium confidence
AA Agentic Index51.0%measured
MMLU-Pro88.5%estimated ± 4.8 pp, medium confidence
AA-IFBench63.5%measured
FailureSensorIQ68.2%estimated ± 1.7 pp, high confidence
GPQA95.0%estimated ± 1.6 pp, high confidence
MMLU97.3%estimated ± 1.2 pp, medium confidence
Vals GPQA Diamond93.2%measured
Vals LiveCodeBench89.8%measured
Vals MMLU-Pro91.5%measured
MMMLU89.7%estimated ± 1.7 pp, medium confidence
Chess Puzzles47.0%measured
AA MMLU-Pro91.0%estimated ± 0.7 pp, medium confidence
Gert Labs72.7%measured
AA-AnalystAgent48.8%measured
AA EnterpriseOps-Gym51.1%measured
AA Harvey LAB v1.093.6%measured
AA Tau3 Banking38.1%measured
AA Terminal-Bench 2.184.6%measured
ApprenticeBench34.0%measured
ARC-AGI-198.5%measured
ARC-AGI-289.2%measured
BabaIsBench100.0%measured
CursorBench 3.170.6%measured
CursorBench 3.270.5%measured
EBR-bench39.5%measured
Terminal-Bench 3.034.1%measured
FrontierCode 1.1 Main53.5%measured
FrontierMath v2 (Tier 4)90.2%measured
FrontierMath v2 (Tiers 1-3)87.0%measured
FrontierSWE v247.0%measured
Furniture Assembly35.8%measured
MaxxBench63.5%measured
MazeBench11.0%measured
MirrorCode70.0%measured
Mystery Game Puzzles52.0%measured
NYT Connections92.7%measured
OfficeQA Pro57.9%measured
OSWorld-Verified85.0%measured
ParseBench-169.8%measured
Terminal-Bench 4.144.9%measured
Terminal-Bench Hard62.9%measured
Vals SWE-bench95.0%measured
Terminal-Bench 2.1 (Vals AI)80.5%measured
VulcanBench v389.5%measured
CWE-Bench-AA60.5%estimated ± 8.1 pp, medium confidence
AA Cyber Index29.6%estimated ± 9.2 pp, low confidence
DeepsecBench-AA29.7%estimated ± 8.2 pp, medium confidence
Harvey LAB-AA v1.17.4%estimated ± 4.0 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)61.0%estimated ± 4.3 pp, medium confidence
HealthBench Professional (raw, Artificial Analysis)69.4%estimated ± 0.8 pp, medium confidence
AA ITBench51.6%estimated ± 3.2 pp, high confidence
Bug Hunt Bench25.4%estimated ± 7.7 pp, medium confidence
Chartography (tools)79.5%estimated ± 4.3 pp, low confidence
CursorBench 4.044.4%estimated ± 2.0 pp, high confidence
CWE-bench v159.5%estimated ± 5.1 pp, medium confidence
DeepSWE72.3%estimated ± 4.3 pp, high confidence
DRACO87.9%estimated ± 0.7 pp, low confidence
FrontierCode 2.029.0%estimated ± 4.8 pp, high confidence
FrontierPhysics1.8%estimated ± 4.4 pp, high confidence
HLE w/o tools54.6%estimated ± 4.9 pp, high confidence
OSWorld 2.019.9%estimated ± 3.6 pp, high confidence
OSWorld 2.177.6%estimated ± 3.6 pp, high confidence
OSWorld 2.1 Offline71.9%estimated ± 1.9 pp, medium confidence
PostTrainBench v1.134.4%estimated ± 3.7 pp, high confidence
ProgramBench89.1%estimated ± 6.0 pp, low confidence
SpireClimb5.8%estimated ± 7.2 pp, medium confidence
StealthMark45.8%estimated ± 6.5 pp, medium confidence
SWE-bench Multimodal56.7%estimated ± 5.6 pp, low confidence
Terminal-Bench 4.048.1%estimated ± 4.5 pp, high confidence
Terminal-Bench-Science 0.133.8%estimated ± 10.7 pp, low confidence
Toolathlon-Verified74.7%estimated ± 2.3 pp, medium confidence
VulcanBench Frontier v490.8%estimated ± 2.3 pp, medium confidence
AA AutomationBench66.6%estimated ± 9.9 pp, low confidence
GDP.pdf25.6%estimated ± 4.2 pp, high confidence
AA Terminal-Bench 4.044.5%estimated ± 6.2 pp, medium confidence
ARC-AGI-311.4%estimated ± 8.6 pp, low confidence
ARC-AGI-3 Provider Adapter97.1%estimated ± 7.3 pp, medium confidence
cursorBench4153.7%estimated ± 3.7 pp, high confidence
CWE-bench60.1%estimated ± 4.6 pp, high confidence
MKA46.6%estimated ± 4.6 pp, high confidence
SkillsBench official leaderboard (BenchFlow)57.1%estimated ± 6.1 pp, medium confidence
BinaryAudit35.8%estimated ± 12.1 pp, low confidence
OTelBench22.5%estimated ± 6.6 pp, medium confidence
Vibe Code Bench81.0%estimated ± 5.6 pp, low confidence
FACTS Multimodal51.8%estimated ± 4.0 pp, high confidence
Hermes Bench72.0%estimated ± 2.9 pp, high confidence
Hermes Index58.8%estimated ± 3.1 pp, high confidence
ParseBench-275.2%estimated ± 4.3 pp, high confidence
CharXiv86.9%estimated ± 5.1 pp, low confidence
CharXiv w/o tools84.7%estimated ± 1.8 pp, medium confidence
ExploitBench32.4%estimated ± 6.5 pp, medium confidence
ExploitGym30.8%estimated ± 2.7 pp, high confidence
C-Eval94.4%estimated ± 1.2 pp, low confidence
Claw-Eval71.9%estimated ± 6.2 pp, low confidence
HMMT Feb 2026100.0%estimated ± 3.1 pp, medium confidence
HMMT Nov 202596.0%estimated ± 1.5 pp, low confidence
MCP Atlas84.7%estimated ± 5.8 pp, low confidence
MMAnswerBench96.0%estimated ± 2.4 pp, low confidence
MMLU-Redux95.8%estimated ± 3.3 pp, medium confidence
MMMU-Pro81.6%estimated ± 6.9 pp, medium confidence
NL2Repo58.9%estimated ± 4.7 pp, high confidence
PostTrainBench v142.9%estimated ± 3.6 pp, medium confidence
ScreenSpot Pro89.2%estimated ± 10.1 pp, low confidence
SuperGPQA100.0%estimated ± 9.3 pp, low confidence
Terminal-Bench 2.081.9%estimated ± 5.4 pp, low confidence
Toolathlon58.9%estimated ± 1.7 pp, low confidence
AA LiveCodeBench100.0%estimated ± 2.6 pp, medium confidence
HealthBench Hard33.8%estimated ± 8.6 pp, low confidence
LiveCodeBench Pro97.2%estimated ± 6.0 pp, low confidence
React Native Evals86.4%estimated ± 2.7 pp, low confidence
ResearchClawBench21.2%estimated ± 1.9 pp, medium confidence
SWE-Rebench61.8%estimated ± 3.6 pp, low confidence
PostTrainBench33.2%estimated ± 2.4 pp, high confidence
DAYJOB1.4%estimated ± 2.7 pp, high confidence
DAYJOB Healthcare0.0%estimated ± 4.0 pp, high confidence
Finance Agent v261.2%estimated ± 2.4 pp, medium confidence
BioMysteryBench (human-solvable)89.5%estimated ± 0.6 pp, medium confidence
FrontierCode 1.1 Extended62.4%estimated ± 1.6 pp, medium confidence
HealthBench (raw)61.8%estimated ± 6.0 pp, medium confidence
HealthBench (length-adjusted)58.4%estimated ± 2.5 pp, low confidence
HealthBench Professional60.5%estimated ± 4.2 pp, medium confidence
HealthBench Professional (raw)71.7%estimated ± 4.5 pp, high confidence
HLE-Verified54.2%estimated ± 1.3 pp, medium confidence
LABBench286.1%estimated ± 1.7 pp, medium confidence
ScarfBench68.9%estimated ± 4.5 pp, low confidence
BlenderWorld (text-only)80.2%estimated ± 2.8 pp, medium confidence
Cookie-Bench88.7%estimated ± 1.3 pp, medium confidence
STEMBench2.7%estimated ± 0.2 pp, medium confidence
SimpleQA34.9%estimated ± 7.8 pp, low confidence
LiveCodeBench87.6%estimated ± 14.6 pp, low confidence
Agents' Last Exam43.4%estimated ± 9.7 pp, medium confidence
OpenHarmony Bench v1.058.9%estimated ± 2.4 pp, medium confidence
SEC-Bench Pro60.3%estimated ± 7.5 pp, medium confidence
MRCRv285.9%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.5%estimated ± 2.6 pp, low confidence
BioBasicsBench67.7%estimated ± 2.8 pp, low confidence
LifeVal79.4%estimated ± 2.3 pp, medium confidence
MMMU-Pro w/ Python83.1%estimated ± 2.7 pp, medium confidence
ExploitBench Internal Port1.2%estimated ± 1.4 pp, medium confidence