Multi-turn Instruction Following on MultiIF (Accuracy)
91.28AccuracyAAPA
Evaluation Results
| Method | Links | |
|---|---|---|
| AAPAParameters=4B2025.09 | 91.28 | |
| Qwen3 ModelParameters=235B-25072025.09 | 90.37 | |
| Qwen3 ModelParameters=8B2025.09 | 86.69 | |
| Qwen3 ModelParameters=32B2025.09 | 85.79 | |
| Qwen3 ModelParameters=4B-25072025.09 | 85.09 | |
| Qwen3 ModelParameters=4B2025.09 | 82 | |
| AAPAParameters=0.6B2025.09 | 76.73 | |
| Qwen3 ModelParameters=1.7B2025.09 | 68.44 | |
| Qwen3 ModelParameters=0.6B2025.09 | 57.6 |