Mathematical Reasoning on DEEPMATH-103K
95.5AccuracyTRS
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TRSModel=Gemini-3-Flash2026.04 | 95.5 | 6,106 | 82.5 | |
| DirectModel=Doubao Seed (Seed-1.8 Pro)2026.04 | 95.1 | 3,453 | 100 | |
| TRSModel=Doubao Seed (Seed-1.8 Pro)2026.04 | 94.9 | 1,411 | 46.2 | |
| DirectModel=Gemini-3-Flash2026.04 | 94.8 | 7,490 | 100 | |
| DirectModel=GPT-OSS-120B2026.04 | 93.7 | 1,257 | 100 | |
| TRSModel=GPT-OSS-120B2026.04 | 93.7 | 976 | 83.1 | |
| TRSModel=GPT-4o-mini2026.04 | 61.4 | 650 | 93.1 | |
| DirectModel=GPT-4o-mini2026.04 | 59.6 | 819 | 100 |