True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise31.9p
#36
5 Top-Tier private tests
Coding & SWE25.6p
#36
5 Top-Tier private tests
Math & Scientific Reasoning23.9p
#38
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 & Enterprise31.9p (5)32.6p (2)+0.7p
Coding & SWE25.6p (5)28.6p (1)+3.0p
General Knowledge & QA49.1p (3)74.2p (1)+25.1p
Math & Scientific Reasoning23.9p (3)33.4p (3)+9.5p
Non-comparable pre-release tests omitted from audited scoring (model released 2026-09-02): SWE-bench Pro V2 (2026-09-22).
Clean Evals Avg (Tiers S & A)
32.6p
16 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
42.2p
7 public or aging tests
Matched Public Lift (Boost)
+14.9p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of Gemini 3.8 Flash

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

Top-Tier Validated Tests (Tiers S & A) (16)

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
AutomationBench-AAPrivate / FreshAgentic & Enterprise#30/5040.8p
MKA (Multimodal Knowledge Agent)Private / FreshAgentic & Enterprise#20/2330.0p
Harvey LAB-AA v1.1Private / FreshAgentic & Enterprise#21/2129.6p
FrontierCode-AA 2.0Private / FreshCoding & SWE#34/4627.8p
FrontierMathPrivate / FreshMath & Scientific Reasoning#25/2726.7p
Code2Video-AAPrivate / FreshCoding & SWE#20/2318.9p
MLCR-AAPrivate / FreshMath & Scientific Reasoning#45/468.4p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#46/508.2p
ARC-AGI-2Private / FreshGeneral Knowledge & QA#11/2262.1p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#21/4757.9p
AA-AnalystAgentPrivate / FreshAgentic & Enterprise#13/2451.1p
AA-3DPrivate / FreshCoding & SWE#18/2236.9p
CritPt v2Private / FreshMath & Scientific Reasoning#24/4036.6p
CyberBench v1.1Private / FreshCoding & SWE#21/2528.6p
LiveBenchPrivate / FreshGeneral Knowledge & QA#25/2827.3p
Code Migration v1.1Private / FreshCoding & SWE#21/2415.7p

Mid-Tier Evaluations (Tier B) (2)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#20/4960.4p
VIALS-AASemi-ExposedMath & Scientific Reasoning#16/2331.8p

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

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
GDP.pdfPublic / AgingAgentic & Enterprise#26/5049.0p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#26/4142.2p
HealthBenchPublic / AgingMath & Scientific Reasoning#16/1633.1p
Terminal-Bench 4.0Public / AgingCoding & SWE#36/5028.6p
BioMysteryBenchPublic / AgingMath & Scientific Reasoning#21/2324.9p
GDPval-AA v2.1Public / AgingAgentic & Enterprise#35/4316.2p
MMMU-ProPublic / AgingGeneral Knowledge & QA#5/2274.2p