Mathematics on AIME 2025 (Score, TPF)
63.33ScoreLLaDA2.1-flash
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LLaDA2.1-flashInference Mode=S Mode2026.02 | 63.33 | 5.36 | |
| LLaDA2.1-flashInference Mode=Q Mode2026.02 | 63.33 | 3.46 | |
| Qwen3-30B-A3B-Inst-25072026.02 | 61.88 | 1 | |
| LLaDA2.0-flash2026.02 | 60 | 4.57 | |
| Ling-flash-2.02026.02 | 55.89 | 1 | |
| Ling-mini-2.02026.02 | 47.66 | — | |
| LLaDA2.1-minimode=Q Mode2026.02 | 43.33 | 3.29 | |
| LLaDA2.0-mini2026.02 | 36.67 | 2.41 | |
| LLaDA2.1-minimode=S Mode2026.02 | 36.67 | 6.34 | |
| Qwen3-8Bno think=true2026.02 | 22.08 | — |