Dynamic Skill Utilization on SkillUse
71.3Skill Utilization (w/ attachment)Claude-4.6-Sonnet-Thinking
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Claude-4.6-Sonnet-ThinkingSize=-2026.06 | 71.3 | 75.9 | 73.6 | |
| Gemini-3.1-Pro-PreviewSize=-2026.06 | 70.9 | 78.1 | 74.5 | |
| Doubao-2.0-ProSize=-2026.06 | 70.3 | 76.9 | 73.6 | |
| DeepSeek-V3.2Size=671B2026.06 | 70.1 | 71.6 | 70.8 | |
| S1-DeepResearchSize=32B2026.06 | 69.7 | 71.7 | 70.1 | |
| Kimi-K2.5Size=1T2026.06 | 64.3 | 73.5 | 68.9 | |
| Qwen3.5-397BSize=397B2026.06 | 63.6 | 68.5 | 66 | |
| GLM-5Size=744B2026.06 | 61.9 | 71.9 | 66.9 | |
| MiniMax-M2.7Size=230B2026.06 | 60.3 | 67.2 | 63.7 | |
| GPT-5.2Size=-2026.06 | 55.6 | 69.7 | 62.7 | |
| Qwen3-235BSize=235B2026.06 | 46 | 46.2 | 46.1 | |
| Qwen3-32BSize=32B2026.06 | 45.6 | 44.3 | 44.9 |