True Index: Domain Rankings
Agentic & Enterprise49.4p
#23
5 Top-Tier private testsCoding & SWE62.7p
#17
5 Top-Tier private testsMath & Scientific Reasoning58.5p
#19
3 Top-Tier private testsHead-to-Head: Clean Evals vs. Legacy Benchmarks
| Capability Domain | Clean Evals (Tiers S & A) | Legacy Benchmarks (Tiers C & D) | Public Lift (Boost) |
|---|---|---|---|
| Agentic & Enterprise | 49.4p | 46.3p | -3.1p |
| Coding & SWE | 62.7p | 39.2p | -23.6p |
| Math & Scientific Reasoning | 58.5p | 67.8p | +9.4p |
Clean Evals Avg (Tiers S & A)
55.3p
15 private, leak-resistant tests
Legacy Benchmarks Avg (Tiers C & D)
51.1p
8 public or aging tests
Matched Public Lift (Boost)
-12.7p
Near zero or negative indicates clean generalization
Machine-Readable Export
Audited Benchmark Performance of Claude Haiku 5.5
Grouped by benchmark integrity so private capability signal stands apart from aging public tests.
Top-Tier Validated Tests (Tiers S & A) (15)
Private, leak-resistant tests that build real capability signal.
| Benchmark | Benchmark Integrity Status | Domain | Rank | Percentile |
|---|---|---|---|---|
| MLCR-AA | Private / Fresh | Math & Scientific Reasoning | #8/46 | 86.0p |
| FrontierCode-AA 2.0 | Private / Fresh | Coding & SWE | #14/46 | 72.9p |
| AA-Briefcase v1.1 | Private / Fresh | Agentic & Enterprise | #16/50 | 69.4p |
| Code2Video-AA | Private / Fresh | Coding & SWE | #7/23 | 66.5p |
| MKA (Multimodal Knowledge Agent) | Private / Fresh | Agentic & Enterprise | #13/23 | 58.8p |
| FrontierMath | Private / Fresh | Math & Scientific Reasoning | #16/27 | 50.4p |
| Harvey LAB-AA v1.1 | Private / Fresh | Agentic & Enterprise | #19/21 | 37.2p |
| AutomationBench-AA | Private / Fresh | Agentic & Enterprise | #48/50 | 4.1p |
| AA-AnalystAgent | Private / Fresh | Agentic & Enterprise | #6/24 | 77.7p |
| CyberBench v1.1 | Private / Fresh | Coding & SWE | #5/25 | 77.7p |
| LiveBench | Private / Fresh | General Knowledge & QA | #17/28 | 54.2p |
| AA-3D | Private / Fresh | Coding & SWE | #15/22 | 48.5p |
| Code Migration v1.1 | Private / Fresh | Coding & SWE | #13/24 | 48.1p |
| AA-Omniscience v1.1 | Private / Fresh | General Knowledge & QA | #26/47 | 47.0p |
| CritPt v2 | Private / Fresh | Math & Scientific Reasoning | #23/40 | 39.0p |
Mid-Tier Evaluations (Tier B) (2)
Partially exposed or transitional evaluations with moderate leakage risk.
| Benchmark | Benchmark Integrity Status | Domain | Rank | Percentile |
|---|---|---|---|---|
| HLE-Diamond | Semi-Exposed | General Knowledge & QA | #22/49 | 56.2p |
| VIALS-AA | Semi-Exposed | Math & Scientific Reasoning | #12/23 | 50.0p |
Legacy & Vulnerable Benchmarks (Tiers C & D) (8)
Public or aging benchmarks prone to overfitting and training-data leaks.
| Benchmark | Benchmark Integrity Status | Domain | Rank | Percentile |
|---|---|---|---|---|
| HealthBench | Public / Aging | Math & Scientific Reasoning | #5/16 | 84.2p |
| BioMysteryBench | Public / Aging | Math & Scientific Reasoning | #12/23 | 60.7p |
| Terminal-Bench-Science 0.1 | Public / Aging | Math & Scientific Reasoning | #19/41 | 58.6p |
| SWE Atlas - Codebase QnA | Public / Aging | Coding & SWE | #11/15 | 47.5p |
| GDPval-AA v2.1 | Public / Aging | Agentic & Enterprise | #22/43 | 46.8p |
| GDP.pdf | Public / Aging | Agentic & Enterprise | #27/50 | 45.9p |
| Terminal-Bench 4.0 | Public / Aging | Coding & SWE | #27/50 | 44.9p |
| SWE-bench Pro V2 | Public / Aging | Coding & SWE | #8/8 | 25.1p |