True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise47.6p
#26
4 Top-Tier private tests
Coding & SWE38.8p
#32
5 Top-Tier private tests
Math & Scientific Reasoning44.2p
#27
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 & Enterprise47.6p (4)61.3p (2)+13.7p
Coding & SWE38.8p (5)59.2p (1)+20.4p
General Knowledge & QA44.8p (1)52.6p (1)+7.8p
Math & Scientific Reasoning44.2p (3)49.7p (1)+5.5p
Clean Evals Avg (Tiers S & A)
43.8p
13 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
55.7p
5 public or aging tests
Matched Public Lift (Boost)
+14.0p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of Qwen3.8 Max (0902)

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

Top-Tier Validated Tests (Tiers S & A) (13)

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#14/5073.5p
MLCR-AAPrivate / FreshMath & Scientific Reasoning#24/4651.4p
FrontierMathPrivate / FreshMath & Scientific Reasoning#16/2750.4p
MKA (Multimodal Knowledge Agent)Private / FreshAgentic & Enterprise#16/2346.4p
FrontierCode-AA 2.0Private / FreshCoding & SWE#28/4641.3p
Code2Video-AAPrivate / FreshCoding & SWE#14/2340.9p
AutomationBench-AAPrivate / FreshAgentic & Enterprise#37/5026.5p
Code Migration v1.1Private / FreshCoding & SWE#10/2460.3p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#27/4744.8p
AA-AnalystAgentPrivate / FreshAgentic & Enterprise#16/2444.0p
AA-3DPrivate / FreshCoding & SWE#19/2233.1p
CritPt v2Private / FreshMath & Scientific Reasoning#26/4030.7p
CyberBench v1.1Private / FreshCoding & SWE#25/2518.2p

Mid-Tier Evaluations (Tier B) (2)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#29/4941.7p
VIALS-AASemi-ExposedMath & Scientific Reasoning#20/2313.6p

Legacy & Vulnerable Benchmarks (Tiers C & D) (5)

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
GDPval-AA v2.1Public / AgingAgentic & Enterprise#14/4365.5p
Terminal-Bench 4.0Public / AgingCoding & SWE#21/5059.2p
GDP.pdfPublic / AgingAgentic & Enterprise#21/5057.1p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#23/4149.7p
MMMU-ProPublic / AgingGeneral Knowledge & QA#10/2252.6p