True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise59.0p
#19
3 Top-Tier private tests
Coding & SWE61.1p
#18
4 Top-Tier private tests
Math & Scientific Reasoning59.1p
#18
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 & Enterprise59.0p (3)55.6p (2)-3.5p
Coding & SWE61.1p (4)53.1p (1)-8.0p
General Knowledge & QA44.7p (3)39.8p (1)-4.9p
Math & Scientific Reasoning59.1p (3)47.2p (2)-11.9p
Clean Evals Avg (Tiers S & A)
56.0p
13 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
48.9p
6 public or aging tests
Matched Public Lift (Boost)
-6.4p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of DeepSeek V4.1 Flash 1017

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
MLCR-AAPrivate / FreshMath & Scientific Reasoning#15/4671.4p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#18/5065.3p
FrontierCode-AA 2.0Private / FreshCoding & SWE#18/4663.9p
AutomationBench-AAPrivate / FreshAgentic & Enterprise#20/5061.2p
MKA (Multimodal Knowledge Agent)Private / FreshAgentic & Enterprise#15/2350.6p
FrontierMathPrivate / FreshMath & Scientific Reasoning#16/2750.4p
Vibe Code Bench 1-100 v1.1Private / FreshCoding & SWE#8/1040.4p
CyberBench v1.1Private / FreshCoding & SWE#7/2571.7p
Code Migration v1.1Private / FreshCoding & SWE#8/2468.4p
LiveBenchPrivate / FreshGeneral Knowledge & QA#15/2860.9p
CritPt v2Private / FreshMath & Scientific Reasoning#16/4055.5p
ARC-AGI-2Private / FreshGeneral Knowledge & QA#15/2243.7p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#34/4729.5p

Mid-Tier Evaluations (Tier B) (2)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#26/4947.9p
VIALS-AASemi-ExposedMath & Scientific Reasoning#17/2327.3p

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#9/4377.4p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#21/4153.6p
Terminal-Bench 4.0Public / AgingCoding & SWE#24/5053.1p
BioMysteryBenchPublic / AgingMath & Scientific Reasoning#17/2340.8p
GDP.pdfPublic / AgingAgentic & Enterprise#33/5033.7p
MMMU-ProPublic / AgingGeneral Knowledge & QA#13/2239.8p