Mathematical Reasoning on HMMT Feb25
94.6Pass@1Nemotron-Cascade-2 30B-A3B
Evaluation Results
| Method | Links | |
|---|---|---|
| Nemotron-Cascade-2 30B-A3B2026.03 | 94.6 | |
| Nemotron-3-Super 120B-A12B2026.03 | 93.7 | |
| Qwen3.5 35B-A3B2026.03 | 89 | |
| Nemotron-3-Nano 30B-A3BOfficial/Recommended Settings=true2026.03 | 84.6 | |
| Qwen3-8Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 44.79 | |
| GPT-5.2-chat (teacher)2026.05 | 43.75 | |
| QUESTA-Nemotron-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 41.67 | |
| ROPDThinking Mode=Thinking2026.05 | 39.16 | |
| T-JudgeThinking Mode=Thinking2026.05 | 38.75 | |
| OVDThinking Mode=Thinking2026.05 | 38.34 | |
| GADThinking Mode=Thinking2026.05 | 35 | |
| Qwen3-4B (student)Thinking Mode=Thinking2026.05 | 33.33 | |
| Qwen3-30B-A3B (teacher)Access=–2026.05 | 33.33 | |
| DeepSeek-R1-Distill-32Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 33 | |
| ROPDThinking Mode=Non-Thinking2026.05 | 31.69 | |
| Nemotron-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 31.56 | |
| OVDThinking Mode=Non-Thinking2026.05 | 29.11 | |
| T-JudgeThinking Mode=Non-Thinking2026.05 | 28.94 | |
| ROPDAccess=text2026.05 | 25.4 | |
| ExOPDAccess=logit2026.05 | 22.42 | |
| Qwen3-1.7Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 22.19 | |
| LOPDAccess=logit2026.05 | 20.42 | |
| GADThinking Mode=Non-Thinking2026.05 | 12.84 | |
| DeepSeek-R1-Distill-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 12 | |
| SFTAccess=text2026.05 | 11.62 | |
| Qwen3-4B (student)Thinking Mode=Non-Thinking2026.05 | 10.42 | |
| Qwen3-4B (student)Access=–2026.05 | 10.42 |