True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise70.4p
#13
5 Top-Tier private tests
Coding & SWE53.9p
#24
6 Top-Tier private tests
Math & Scientific Reasoning34.7p
#33
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 & Enterprise70.4p (5)57.8p (2)-12.6p
Coding & SWE53.9p (6)37.8p (1)-16.1p
Math & Scientific Reasoning34.7p (3)53.9p (3)+19.2p
Clean Evals Avg (Tiers S & A)
53.3p
17 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
49.8p
6 public or aging tests
Matched Public Lift (Boost)
-2.7p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of Grok 4.7

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
Harvey LAB-AA v1.1Private / FreshAgentic & Enterprise#5/2188.2p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#11/5079.6p
FrontierCode-AA 2.0Private / FreshCoding & SWE#17/4666.1p
Code2Video-AAPrivate / FreshCoding & SWE#8/2362.9p
AutomationBench-AAPrivate / FreshAgentic & Enterprise#22/5057.1p
MKA (Multimodal Knowledge Agent)Private / FreshAgentic & Enterprise#14/2354.6p
MLCR-AAPrivate / FreshMath & Scientific Reasoning#24/4651.4p
Vibe Code Bench 1-100 v1.1Private / FreshCoding & SWE#7/1049.6p
FrontierMathPrivate / FreshMath & Scientific Reasoning#26/2721.9p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#12/4777.6p
AA-AnalystAgentPrivate / FreshAgentic & Enterprise#8/2472.4p
Code Migration v1.1Private / FreshCoding & SWE#12/2452.2p
LiveBenchPrivate / FreshGeneral Knowledge & QA#18/2850.8p
CyberBench v1.1Private / FreshCoding & SWE#15/2548.0p
AA-3DPrivate / FreshCoding & SWE#16/2244.7p
ARC-AGI-2Private / FreshGeneral Knowledge & QA#17/2234.6p
CritPt v2Private / FreshMath & Scientific Reasoning#26/4030.7p

Mid-Tier Evaluations (Tier B) (2)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#30/4939.6p
VIALS-AASemi-ExposedMath & Scientific Reasoning#19/2318.2p

Legacy & Vulnerable Benchmarks (Tiers C & D) (6)

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
GDPval-AA v2.1Public / AgingAgentic & Enterprise#11/4372.6p
HealthBenchPublic / AgingMath & Scientific Reasoning#9/1665.6p
BioMysteryBenchPublic / AgingMath & Scientific Reasoning#14/2348.8p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#24/4147.2p
GDP.pdfPublic / AgingAgentic & Enterprise#29/5042.9p
Terminal-Bench 4.0Public / AgingCoding & SWE#31/5037.8p