Scientific Reasoning on GPQA Diamond (Pass Ratio)
60.5Pass RatioREASONING ARENA-Live
Evaluation results
| Method | Links | |
|---|---|---|
| REASONING ARENA-Livetype=adaptive routing2026.06 | 60.5 | |
| ArenaRLtype=judge-only2026.06 | 59.6 | |
| REASONING ARENAtype=adaptive routing2026.06 | 59.5 | |
| Adaptive Pointwisetype=adaptive routing2026.06 | 58.5 | |
| RLAIFtype=judge-only2026.06 | 56.9 | |
| RLVRtype=verifier-only RL2026.06 | 54.8 |