Reasoning on bbh-zh
87.52Overall ScoreLLaDA2.0-flash
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LLaDA2.0-flash2026.02 | 87.52 | 321 | |
| LLaDA2.1-flashInference Mode=Q Mode2026.02 | 86.23 | 377 | |
| Qwen3-30B-A3B-Inst-25072026.02 | 86.18 | 100 | |
| LLaDA2.1-flashInference Mode=S Mode2026.02 | 82.55 | 578 | |
| Qwen3-8Bno think=true2026.02 | 80.09 | — | |
| LLaDA2.0-mini2026.02 | 75.75 | 2.77 | |
| Ling-flash-2.02026.02 | 75.09 | 100 | |
| LLaDA2.1-minimode=Q Mode2026.02 | 70.4 | 2.35 | |
| LLaDA2.1-minimode=S Mode2026.02 | 67.65 | 3.89 | |
| Ling-mini-2.02026.02 | 66.11 | — |