OverviewCompare models
Compare models
Compare exact releases using results from the same evaluator.
Selected benchmark
Compare the two releases within one evaluator and metric.
- 59.60%
Bars start at zero. 1 reported results for the selected models; ranks apply to this selection only; ties share a rank. Published effort settings may vary. Source ↗
Beam: no published result in this group. No score is inferred.
Across the tests
Each row uses one evaluator and scale. Percentage differences are in percentage points (pp).
| Test / evaluator | Beam | Claude Opus 5.5 | DifferenceModel 1 − model 2 |
|---|---|---|---|
| ITBench-AAaa% · Higher is better | No result | 38.23 | — |
| MLCR-AAaa% · Higher is better | No result | 66.67 | — |
| Vals RSI Indexvals% · Higher is better | No result | 37.31 | — |
| Vibe Code Bench 1–100vals% · Higher is better | No result | 30.36 | — |
| CUA-benchvals% · Higher is better | No result | 14.00 | — |
| Harvey LAB-AAaa% · Higher is better | No result | 91.20 | — |
| AutomationBench-AAaa% · Higher is better | No result | 69.54 | — |
| SciCodeaa% · Higher is better | No result | 66.90 | — |
| GDP.pdfaa% · Higher is better | No result | 26.20 | — |
| CritPtaa% · Higher is better | No result | 31.71 | — |
| AA-Omniscienceaascore · Higher is better | No result | 46.42 | — |
| AA-LCR v1.1aa% · Higher is better | No result | 84.67 | — |
| Terminal-Bench 4.0aa% · Higher is better | No result | 59.60 | — |
| Terminal-Bench 4.0vals% · Higher is better | No result | 65.15 | — |
| Terminal-Bench 2.1vals% · Higher is better | No result | 87.64 | — |
| MMMU Proaa% · Higher is better | No result | 87.69 | — |
| Vibe Code Bench v1.1vals% · Higher is better | No result | 90.29 | — |
| Terminal-Bench Science 0.1aa% · Higher is better | No result | 59.05 | — |
| Terminal-Bench Science 0.1vals% · Higher is better | No result | 47.14 | — |
| ProgramBenchvals% · Higher is better | No result | 18.50 | — |
| ProofBench v1.1vals% · Higher is better | No result | 100.00 | — |
| Harvey's Legal Agent Benchmarkvals% · Higher is better | No result | 3.75 | — |
| MysteryMechanismvals% · Higher is better | No result | 49.55 | — |
| Vals Indexvals% · Higher is better | No result | 66.97 | — |
| BioMysteryBenchvals% · Higher is better | No result | 79.26 | — |
| Code Migrationvals% · Higher is better | No result | 66.65 | — |
| CyberBench v1.1vals% · Higher is better | No result | 55.36 | — |
| EMBvals% · Higher is better | No result | 75.94 | — |
| Finance Agent (v2)vals% · Higher is better | No result | 58.59 | — |
| IOIvals% · Higher is better | No result | 95.06 | — |
| Legal Research Benchvals% · Higher is better | No result | 50.48 | — |
| MedCodevals% · Higher is better | No result | 49.80 | — |
| MedScribevals% · Higher is better | No result | 91.43 | — |
| Public Benefits Bench v1.1vals% · Higher is better | No result | 70.64 | — |
| SAGEvals% · Higher is better | No result | 45.83 | — |
| SRE Benchvals% · Higher is better | No result | 33.59 | — |
| Tax Agent Benchvals% · Higher is better | No result | 70.50 | — |
| FrontierCode v1.1aa% · Higher is better | No result | 52.17 | — |
| GDPval-AA v2.1aaElo · Higher is better | No result | 1,846.17 | — |
| AA-Briefcase v1.1aaElo · Higher is better | No result | 1,821.85 | — |
| Humanity's Last Examaa% · Higher is better | No result | 61.35 | — |
| Artificial Analysis Intelligence Indexaaindex · Higher is better | No result | 58 | — |