Mathematical Reasoning on AIME 2025 (Pass@1, BoN, KT, SC, WSC, JC)
69.9Pass@1 AccuracyPass@1
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Pass@1N=24, Backbone=GPT-OSS-20B2026.05 | 69.9 | — | — | — | — | — | |
| Pass@1N=48, Backbone=GPT-OSS-20B2026.05 | 69.9 | — | — | — | — | — | |
| Pass@1N=120, Backbone=GPT-OSS-20B2026.05 | 69.8 | — | — | — | — | — | |
| BoNN=24, Backbone=GPT-OSS-20B2026.05 | — | 91.2 | — | — | — | — | |
| BoNN=48, Backbone=GPT-OSS-20B2026.05 | — | 91.1 | — | — | — | — | |
| BoNN=120, Backbone=GPT-OSS-20B2026.05 | — | 89.9 | — | — | — | — | |
| JCN=24, Backbone=GPT-OSS-20B2026.05 | — | — | — | — | — | 95.5 | |
| JCN=48, Backbone=GPT-OSS-20B2026.05 | — | — | — | — | — | 96.1 | |
| JCN=120, Backbone=GPT-OSS-20B2026.05 | — | — | — | — | — | 96.6 | |
| KTN=24, Backbone=GPT-OSS-20B2026.05 | — | — | 46.9 | — | — | — | |
| KTN=48, Backbone=GPT-OSS-20B2026.05 | — | — | 62.3 | — | — | — | |
| KTN=120, Backbone=GPT-OSS-20B2026.05 | — | — | 74.3 | — | — | — | |
| SCN=24, Backbone=GPT-OSS-20B2026.05 | — | — | — | 93.5 | — | — | |
| SCN=48, Backbone=GPT-OSS-20B2026.05 | — | — | — | 94.1 | — | — | |
| SCN=120, Backbone=GPT-OSS-20B2026.05 | — | — | — | 94.3 | — | — | |
| WSCN=24, Backbone=GPT-OSS-20B2026.05 | — | — | — | — | 95.3 | — | |
| WSCN=48, Backbone=GPT-OSS-20B2026.05 | — | — | — | — | 96 | — | |
| WSCN=120, Backbone=GPT-OSS-20B2026.05 | — | — | — | — | 96.6 | — |