True Index: Domain Rankings

Private Top-Tier (S & A) standing across 3 core domains
Agentic & Enterprise16.4p
#40
2 Top-Tier private tests
Coding & SWE9.6p
#41
2 Top-Tier private tests
Math & Scientific Reasoning22.9p
#39
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 & Enterprise16.4p (2)24.0p (2)+7.6p
Coding & SWE9.6p (2)23.5p (2)+13.9p
General Knowledge & QA34.1p (3)44.0p (1)+9.9p
Math & Scientific Reasoning22.9p (3)17.1p (1)-5.8p
Clean Evals Avg (Tiers S & A)
20.8p
10 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
27.1p
6 public or aging tests
Matched Public Lift (Boost)
+6.5p
Near zero or negative indicates clean generalization
Machine-Readable Export

Audited Benchmark Performance of Muse Spark Open

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
FrontierMathPrivate / FreshMath & Scientific Reasoning#21/2737.8p
AutomationBench-AAPrivate / FreshAgentic & Enterprise#41/5018.4p
AA-Briefcase v1.1Private / FreshAgentic & Enterprise#43/5014.3p
FrontierCode-AA 2.0Private / FreshCoding & SWE#43/467.5p
MLCR-AAPrivate / FreshMath & Scientific Reasoning#46/466.3p
AA-Omniscience v1.1Private / FreshGeneral Knowledge & QA#18/4764.4p
CritPt v2Private / FreshMath & Scientific Reasoning#29/4024.7p
ARC-AGI-2Private / FreshGeneral Knowledge & QA#20/2220.8p
LiveBenchPrivate / FreshGeneral Knowledge & QA#28/2817.2p
Code Migration v1.1Private / FreshCoding & SWE#22/2411.7p

Mid-Tier Evaluations (Tier B) (1)

Partially exposed or transitional evaluations with moderate leakage risk.

BenchmarkBenchmark Integrity StatusDomainRankPercentile
HLE-DiamondSemi-ExposedGeneral Knowledge & QA#25/4950.0p

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

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

BenchmarkBenchmark Integrity StatusDomainRankPercentile
GDP.pdfPublic / AgingAgentic & Enterprise#31/5038.8p
SWE-bench Pro V2Public / AgingCoding & SWE#7/834.8p
Terminal-Bench-Science 0.1Public / AgingMath & Scientific Reasoning#36/4117.1p
Terminal-Bench 4.0Public / AgingCoding & SWE#44/5012.2p
GDPval-AA v2.1Public / AgingAgentic & Enterprise#38/439.2p
MMMU-ProPublic / AgingGeneral Knowledge & QA#12/2244.0p