Aggregate on Overall Benchmarks
49.6Overall AverageUM-190k
Evaluation Results
| Method | Links | |
|---|---|---|
| UM-190kBackbone=Apertus-8B-SFT2025.11 | 49.6 | |
| UM-187kBackbone=Apertus-8B-SFT2025.11 | 49.17 | |
| UM-170kBackbone=Apertus-8B-SFT2025.11 | 47.95 | |
| TuluDPOBackbone=Apertus-8B-SFT2025.11 | 47.66 | |
| UltraFBBackbone=Apertus-8B-SFT2025.11 | 46.66 | |
| ORPOBackbone=Apertus-8B-SFT2025.11 | 46.45 | |
| HelpSteerBackbone=Apertus-8B-SFT2025.11 | 45.38 | |
| CodePrefBackbone=Apertus-8B-SFT2025.11 | 45.06 | |
| SFTBackbone=Apertus-8B-SFT2025.11 | 44.9 |