True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise47.3p
#27
3 Top-Tier private tests
Coding & SWE77.8p
#10
3 Top-Tier private tests
Math & Scientific Reasoning80.4p
#9
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 & Enterprise47.3p (3)47.1p (2)-0.1p
Coding & SWE77.8p (3)65.3p (1)-12.5p
General Knowledge & QA60.1p (2)57.0p (1)-3.1p
Math & Scientific Reasoning80.4p (3)62.6p (3)-17.9p
Clean Evals Avg (Tiers S & A)
66.4p
11 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
58.0p
7 public or aging tests
Matched Public Lift (Boost)
-6.6p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of GPT-6 Sol

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

Top-Tier Validated Tests (Tiers S & A) (11)

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
MLCR-AAPrivate / FreshMath & Scientific Reasoning#9/4683.9p
FrontierMathPrivate / FreshMath & Scientific Reasoning#8/2780.5p
FrontierCode-AA 2.0Private / FreshCoding & SWE#13/4675.1p
Harvey LAB-AA v1.1Private / FreshAgentic & Enterprise#14/2156.1p
AutomationBench-AAPrivate / FreshAgentic & Enterprise#27/5046.9p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#31/5038.8p
CyberBench v1.1Private / FreshCoding & SWE#2/2586.6p
CritPt v2Private / FreshMath & Scientific Reasoning#7/4076.9p
AA-3DPrivate / FreshCoding & SWE#9/2271.7p
ARC-AGI-2Private / FreshGeneral Knowledge & QA#10/2266.7p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#23/4753.5p

Mid-Tier Evaluations (Tier B) (1)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#15/4970.8p

Legacy & Vulnerable Benchmarks (Tiers C & D) (7)

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
BioMysteryBenchPublic / AgingMath & Scientific Reasoning#10/2368.7p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#16/4167.3p
GDP.pdfPublic / AgingAgentic & Enterprise#17/5066.3p
Terminal-Bench 4.0Public / AgingCoding & SWE#18/5065.3p
HealthBenchPublic / AgingMath & Scientific Reasoning#12/1651.7p
GDPval-AA v2.1Public / AgingAgentic & Enterprise#30/4328.0p
MMMU-ProPublic / AgingGeneral Knowledge & QA#9/2257.0p