Knowledge & Instruction Following on IFEval (Accuracy)
91.31AccuracyQwen3.5-9B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3.5-9BParams=9B, Sampling mode=AR2026.06 | 91.31 | |
| Qwen3.5Params=4B, Sampling mode=AR2026.06 | 90.02 | |
| LLaDA-2.1-flashParams=100B-A5B, Sampling mode=Diffusion2026.06 | 83.36 | |
| LLaDA-2.0-flashParams=100B-A5B, Sampling mode=Diffusion2026.06 | 81.7 | |
| LLaDA-2.1-miniParams=16B-A1B, Sampling mode=Diffusion2026.06 | 81.33 | |
| LLaDA-2.0-miniParams=16B-A1B, Sampling mode=Diffusion2026.06 | 80.78 | |
| Qwen3.5Params=2B, Sampling mode=AR2026.06 | 79.48 | |
| FLAREParams=4B, Sampling mode=Diffusion-Trust2026.06 | 73.57 | |
| FLAREParams=4B, Sampling mode=AR-Trust2026.06 | 73.2 | |
| FLARE-9BParams=9B, Sampling mode=AR-Trust2026.06 | 71.35 | |
| FLAREParams=2B, Sampling mode=AR-Trust2026.06 | 68.95 | |
| FLARE-9BParams=9B, Sampling mode=Diffusion-Trust2026.06 | 63.22 | |
| FLAREParams=2B, Sampling mode=Diffusion-Trust2026.06 | 62.66 | |
| SDARParams=8B, Sampling mode=Diffusion2026.06 | 61.4 | |
| SDAR 30B-A3BParams=30B-A3B, Sampling mode=Diffusion2026.06 | 60.6 | |
| SDARParams=4B, Sampling mode=Diffusion2026.06 | 56.6 | |
| SDARParams=1.7B, Sampling mode=Diffusion2026.06 | 43.4 |