True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise68.0p
#16
5 Top-Tier private tests
Coding & SWE71.7p
#12
5 Top-Tier private tests
Math & Scientific Reasoning53.2p
#22
3 Top-Tier private tests

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

3 matched domains · positive boost means higher legacy scores
Capability Domain Clean Evals (Tiers S & A) Legacy Benchmarks (Tiers C & D) Public Lift (Boost)
Agentic & Enterprise68.0p (5)52.3p (2)-15.7p
Coding & SWE71.7p (5)86.5p (2)+14.9p
Math & Scientific Reasoning53.2p (3)60.4p (4)+7.3p
Clean Evals Avg (Tiers S & A)
67.9p
16 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
66.4p
8 public or aging tests
Matched Public Lift (Boost)
+5.2p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of Gemini 4 Argon

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#3/5095.9p
MKA (Multimodal Knowledge Agent)Private / FreshAgentic & Enterprise#7/2383.3p
Code2Video-AAPrivate / FreshCoding & SWE#6/2370.2p
FrontierMathPrivate / FreshMath & Scientific Reasoning#12/2767.8p
Harvey LAB-AA v1.1Private / FreshAgentic & Enterprise#13/2159.9p
FrontierCode-AA 2.0Private / FreshCoding & SWE#23/4652.6p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#29/5042.9p
MLCR-AAPrivate / FreshMath & Scientific Reasoning#37/4624.2p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#8/4786.3p
CyberBench v1.1Private / FreshCoding & SWE#3/2583.7p
AA-3DPrivate / FreshCoding & SWE#7/2279.4p
ARC-AGI-2Private / FreshGeneral Knowledge & QA#8/2275.9p
LiveBenchPrivate / FreshGeneral Knowledge & QA#11/2874.4p
Code Migration v1.1Private / FreshCoding & SWE#7/2472.4p
CritPt v2Private / FreshMath & Scientific Reasoning#11/4067.5p
AA-AnalystAgentPrivate / FreshAgentic & Enterprise#12/2458.2p

Mid-Tier Evaluations (Tier B) (2)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#11/4979.2p
VIALS-AASemi-ExposedMath & Scientific Reasoning#8/2365.9p

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

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
SWE-bench Pro V2Public / AgingCoding & SWE#1/892.5p
Terminal-Bench 4.0Public / AgingCoding & SWE#10/5080.6p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#11/4179.9p
BioMysteryBenchPublic / AgingMath & Scientific Reasoning#8/2376.7p
GDP.pdfPublic / AgingAgentic & Enterprise#24/5053.1p
GDPval-AA v2.1Public / AgingAgentic & Enterprise#20/4351.5p
Autoform-HardPublic / AgingMath & Scientific Reasoning#6/847.4p
HealthBenchPublic / AgingMath & Scientific Reasoning#15/1637.8p