Reasoning on AIME 2025 (Score, Tool-integrated Score)
92.4AIME ScoreNemotron-Cascade-2-30B-A3B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Nemotron-Cascade-2-30B-A3BContext Length=1M, Evaluation Setting=Official, Tool-integrated Reasoning=false2026.07 | 92.4 | — | |
| Qwen3.5 35B-A3BContext Length=256K, Evaluation Setting=Recommended Settings2026.07 | 91.9 | — | |
| Audex 30B-A3BContext Length=1M, Evaluation Setting=Official, Tool-integrated Reasoning=false2026.07 | 91.2 | — | |
| Qwen3 30B-A3B Thinking-2507Context Length=256K, Evaluation Setting=Official2026.07 | 85 | — | |
| Qwen3-Omni 30B-A3B ThinkingContext Length=64K, Evaluation Setting=Official2026.07 | 73.7 | — | |
| Step-Audio R1.1 33BContext Length=64K, Evaluation Setting=Recommended Settings2026.07 | 44.8 | — | |
| Audex 30B-A3BContext Length=1M, Evaluation Setting=Official, Tool-integrated Reasoning=true2026.07 | — | 98.3 | |
| Nemotron-Cascade-2-30B-A3BContext Length=1M, Evaluation Setting=Official, Tool-integrated Reasoning=true2026.07 | — | 98.6 |