Instruction Following on IFEval (Accuracy (%))
94.3Accuracy (%)Qwen3.6 Plus
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3.6 PlusParams=N/A2026.06 | 94.3 | |
| Kimi K2.5Params=1T2026.06 | 93.9 | |
| VibeThinker-3BParams=3B2026.06 | 93.4 | |
| Qwen3.5-397B-A17BParams=397B2026.06 | 92.6 | |
| DeepSeek V3.2Params=671B2026.06 | 92.6 | |
| GLM-5Params=744B2026.06 | 92.6 | |
| OpenAI o3 (high)Params=N/A2026.06 | 92.1 | |
| Gemini 2.5 ProParams=N/A2026.06 | 90.8 | |
| Gemini 2.5 FlashParams=N/A2026.06 | 89.8 | |
| GPT-OSS (high)Params=120B2026.06 | 89.5 | |
| Grok 4Params=N/A2026.06 | 88 | |
| DeepSeek R1 0528Params=671B2026.06 | 79.1 |