True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise49.0p
#24
2 Top-Tier private tests
Coding & SWE43.6p
#31
5 Top-Tier private tests
Math & Scientific Reasoning13.4p
#42
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 & Enterprise49.0p (2)24.2p (2)-24.8p
Coding & SWE43.6p (5)30.6p (1)-13.0p
General Knowledge & QA5.3p (1)1.0p (1)-4.3p
Math & Scientific Reasoning13.4p (2)38.0p (2)+24.6p
Clean Evals Avg (Tiers S & A)
27.8p
10 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
23.5p
6 public or aging tests
Matched Public Lift (Boost)
-3.6p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of MiMo-V2.6-Flash

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

Top-Tier Validated Tests (Tiers S & A) (10)

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
Code2Video-AAPrivate / FreshCoding & SWE#10/2355.6p
AutomationBench-AAPrivate / FreshAgentic & Enterprise#24/5053.1p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#28/5044.9p
FrontierCode-AA 2.0Private / FreshCoding & SWE#30/4636.8p
Vibe Code Bench 1-100 v1.1Private / FreshCoding & SWE#9/1031.2p
MLCR-AAPrivate / FreshMath & Scientific Reasoning#43/4611.6p
CyberBench v1.1Private / FreshCoding & SWE#6/2574.7p
Code Migration v1.1Private / FreshCoding & SWE#20/2419.8p
CritPt v2Private / FreshMath & Scientific Reasoning#33/4015.2p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#45/475.3p

Mid-Tier Evaluations (Tier B) (1)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#45/498.3p

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

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
BioMysteryBenchPublic / AgingMath & Scientific Reasoning#14/2348.8p
GDPval-AA v2.1Public / AgingAgentic & Enterprise#23/4344.4p
Terminal-Bench 4.0Public / AgingCoding & SWE#35/5030.6p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#31/4127.1p
GDP.pdfPublic / AgingAgentic & Enterprise#48/504.1p
MMMU-ProPublic / AgingGeneral Knowledge & QA#22/221.0p