Mathematical Reasoning on AIME 24 (Rank)
1RankQwen3-30B-A3B-Thinking-2507
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-30B-A3B-Thinking-2507Confidence Interval (CI)=w/ CI2025.10 | 1 | |
| Qwen3-30B-A3B-Thinking-2507Confidence Interval (CI)=w/o CI2025.10 | 1 | |
| Qwen3-4B-Thinking-2507Confidence Interval (CI)=w/ CI2025.10 | 2 | |
| Qwen3-4B-Thinking-2507Confidence Interval (CI)=w/o CI2025.10 | 2 | |
| gpt-oss-20b-highConfidence Interval (CI)=w/ CI2025.10 | 3 | |
| gpt-oss-20b-mediumConfidence Interval (CI)=w/ CI2025.10 | 3 | |
| gpt-oss-20b-mediumConfidence Interval (CI)=w/o CI2025.10 | 3 | |
| Phi-4-reasoning-plusConfidence Interval (CI)=w/ CI2025.10 | 3 | |
| Phi-4-reasoning-plusConfidence Interval (CI)=w/o CI2025.10 | 4 | |
| OpenThinker2-32BConfidence Interval (CI)=w/ CI2025.10 | 4 | |
| Light-R1-14B-DSConfidence Interval (CI)=w/ CI2025.10 | 4 | |
| FuseO1-DeepSeekR1-QwQ-SkyT1-Flash-32BConfidence Interval (CI)=w/ CI2025.10 | 4 | |
| gpt-oss-20b-highConfidence Interval (CI)=w/o CI2025.10 | 5 | |
| AceReason-Nemotron-1.1-7BConfidence Interval (CI)=w/ CI2025.10 | 5 | |
| Phi-4-reasoningConfidence Interval (CI)=w/ CI2025.10 | 5 | |
| gpt-oss-20b-lowConfidence Interval (CI)=w/ CI2025.10 | 6 | |
| Light-R1-14B-DSConfidence Interval (CI)=w/o CI2025.10 | 6 | |
| NVIDIA-Nemotron-Nano-9B-v2Confidence Interval (CI)=w/ CI2025.10 | 6 | |
| FuseO1-DeepSeekR1-QwQ-SkyT1-Flash-32BConfidence Interval (CI)=w/o CI2025.10 | 7 | |
| LIMO-v2Confidence Interval (CI)=w/ CI2025.10 | 7 | |
| OpenThinker2-32BConfidence Interval (CI)=w/o CI2025.10 | 8 | |
| EXAONE-4.0-1.2BConfidence Interval (CI)=w/ CI2025.10 | 8 | |
| AceReason-Nemotron-1.1-7BConfidence Interval (CI)=w/o CI2025.10 | 9 | |
| OpenR1-Distill-7BConfidence Interval (CI)=w/ CI2025.10 | 9 | |
| Phi-4-reasoningConfidence Interval (CI)=w/o CI2025.10 | 10 | |
| OpenThinker3-1.5BConfidence Interval (CI)=w/ CI2025.10 | 10 | |
| NVIDIA-Nemotron-Nano-9B-v2Confidence Interval (CI)=w/o CI2025.10 | 11 | |
| OpenReasoning-Nemotron-1.5BConfidence Interval (CI)=w/ CI2025.10 | 11 | |
| gpt-oss-20b-lowConfidence Interval (CI)=w/o CI2025.10 | 12 | |
| DeepSeek-R1-Distill-Qwen-1.5BConfidence Interval (CI)=w/ CI2025.10 | 12 | |
| Sky-T1-32B-FlashConfidence Interval (CI)=w/ CI2025.10 | 12 | |
| LIMO-v2Confidence Interval (CI)=w/o CI2025.10 | 13 | |
| Bespoke-Stratos-7BConfidence Interval (CI)=w/ CI2025.10 | 13 | |
| EXAONE-4.0-1.2BConfidence Interval (CI)=w/o CI2025.10 | 14 | |
| OpenR1-Distill-7BConfidence Interval (CI)=w/o CI2025.10 | 15 | |
| OpenThinker3-1.5BConfidence Interval (CI)=w/o CI2025.10 | 16 | |
| OpenReasoning-Nemotron-1.5BConfidence Interval (CI)=w/o CI2025.10 | 17 | |
| Sky-T1-32B-FlashConfidence Interval (CI)=w/o CI2025.10 | 18 | |
| DeepSeek-R1-Distill-Qwen-1.5BConfidence Interval (CI)=w/o CI2025.10 | 19 | |
| Bespoke-Stratos-7BConfidence Interval (CI)=w/o CI2025.10 | 20 |