True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise48.8p
#25
3 Top-Tier private tests
Coding & SWE27.1p
#34
2 Top-Tier private tests
Math & Scientific Reasoning42.4p
#30
2 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 & Enterprise48.8p (3)34.6p (2)-14.2p
Coding & SWE27.1p (2)50.7p (2)+23.6p
General Knowledge & QA26.4p (2)26.9p (1)+0.5p
Math & Scientific Reasoning42.4p (2)24.6p (2)-17.8p
Clean Evals Avg (Tiers S & A)
36.2p
9 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
34.2p
7 public or aging tests
Matched Public Lift (Boost)
+0.3p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of MiniMax-M3.1-Flash

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

Top-Tier Validated Tests (Tiers S & A) (9)

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
AutomationBench-AAPrivate / FreshAgentic & Enterprise#9/5083.7p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#30/5040.8p
FrontierCode-AA 2.0Private / FreshCoding & SWE#32/4632.3p
MLCR-AAPrivate / FreshMath & Scientific Reasoning#34/4631.5p
Vibe Code Bench 1-100 v1.1Private / FreshCoding & SWE#10/1022.0p
MKA (Multimodal Knowledge Agent)Private / FreshAgentic & Enterprise#22/2321.8p
CritPt v2Private / FreshMath & Scientific Reasoning#17/4053.2p
LiveBenchPrivate / FreshGeneral Knowledge & QA#21/2840.7p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#42/4712.0p

Mid-Tier Evaluations (Tier B) (2)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#24/4952.1p
VIALS-AASemi-ExposedMath & Scientific Reasoning#21/239.1p

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

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
SWE-bench Pro V2Public / AgingCoding & SWE#4/863.6p
GDP.pdfPublic / AgingAgentic & Enterprise#27/5045.9p
Terminal-Bench 4.0Public / AgingCoding & SWE#31/5037.8p
Autoform-HardPublic / AgingMath & Scientific Reasoning#7/827.0p
GDPval-AA v2.1Public / AgingAgentic & Enterprise#32/4323.3p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#34/4122.1p
MMMU-ProPublic / AgingGeneral Knowledge & QA#16/2226.9p