True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise83.2p
#6
5 Top-Tier private tests
Coding & SWE79.1p
#8
6 Top-Tier private tests
Math & Scientific Reasoning81.4p
#8
3 Top-Tier private tests

Head-to-Head: Clean Evals vs. Legacy Benchmarks

4 matched domains · positive boost means higher legacy scores
Capability Domain Clean Evals (Tiers S & A) Legacy Benchmarks (Tiers C & D) Public Lift (Boost)
Agentic & Enterprise83.2p (5)84.8p (2)+1.6p
Coding & SWE79.1p (6)92.1p (2)+13.0p
General Knowledge & QA88.2p (3)87.1p (1)-1.1p
Math & Scientific Reasoning81.4p (3)82.2p (3)+0.8p
Clean Evals Avg (Tiers S & A)
83.0p
17 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
86.6p
8 public or aging tests
Matched Public Lift (Boost)
+11.2p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of Claude Opus 5.5

Grouped by benchmark integrity so private capability signal stands apart from aging public tests.

Top-Tier Validated Tests (Tiers S & A) (17)

Private, leak-resistant tests that build real capability signal.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
FrontierCode-AA 2.0Private / FreshCoding & SWE#2/46100.0p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#2/5098.0p
MKA (Multimodal Knowledge Agent)Private / FreshAgentic & Enterprise#4/2395.6p
Code2Video-AAPrivate / FreshCoding & SWE#1/2388.5p
FrontierMathPrivate / FreshMath & Scientific Reasoning#6/2786.8p
Vibe Code Bench 1-100 v1.1Private / FreshCoding & SWE#4/1077.2p
AutomationBench-AAPrivate / FreshAgentic & Enterprise#13/5075.5p
MLCR-AAPrivate / FreshMath & Scientific Reasoning#14/4673.4p
Harvey LAB-AA v1.1Private / FreshAgentic & Enterprise#12/2163.7p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#5/4792.9p
LiveBenchPrivate / FreshGeneral Knowledge & QA#6/2891.2p
Code Migration v1.1Private / FreshCoding & SWE#3/2488.5p
AA-3DPrivate / FreshCoding & SWE#5/2287.2p
CritPt v2Private / FreshMath & Scientific Reasoning#3/4084.0p
AA-AnalystAgentPrivate / FreshAgentic & Enterprise#5/2483.0p
ARC-AGI-2Private / FreshGeneral Knowledge & QA#7/2280.5p
CyberBench v1.1Private / FreshCoding & SWE#20/2533.0p

Mid-Tier Evaluations (Tier B) (2)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#5/4991.7p
VIALS-AASemi-ExposedMath & Scientific Reasoning#6/2377.3p

Legacy & Vulnerable Benchmarks (Tiers C & D) (8)

Public or aging benchmarks prone to overfitting and training-data leaks.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
SWE Atlas - Codebase QnAPublic / AgingCoding & SWE#3/1597.5p
GDPval-AA v2.1Public / AgingAgentic & Enterprise#1/4396.1p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#7/4189.9p
Terminal-Bench 4.0Public / AgingCoding & SWE#7/5086.7p
BioMysteryBenchPublic / AgingMath & Scientific Reasoning#5/2386.6p
GDP.pdfPublic / AgingAgentic & Enterprise#13/5073.5p
HealthBenchPublic / AgingMath & Scientific Reasoning#8/1670.2p
MMMU-ProPublic / AgingGeneral Knowledge & QA#2/2287.1p