App-based Task Execution on AppWorld-Challenge
52.8Task Goal Completion (TGC)Trace2Skill
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Trace2SkillSkill Generation=GPT-5.4, Inference Model=Qwen3.5-122B-A10B2026.05 | 52.8 | 36.7 | |
| MIND-SkillSkill Generation=Qwen3.5-122B-A10B, Inference Model=Qwen3.5-122B-A10B2026.05 | 51.8 | 39.6 | |
| ACESkill Generation=Qwen3.5-122B-A10B, Inference Model=Qwen3.5-122B-A10B2026.05 | 51.1 | 34.5 | |
| MIND-SkillSkill Generation=GPT-5.4, Inference Model=Qwen3.5-122B-A10B2026.05 | 50.6 | 37.4 | |
| Skill-extractSkill Generation=GPT-5.4, Inference Model=Qwen3.5-122B-A10B2026.05 | 47.7 | 33.1 | |
| ACESkill Generation=GPT-5.4, Inference Model=Qwen3.5-122B-A10B2026.05 | 47.2 | 33.8 | |
| Trace2SkillSkill Generation=Qwen3.5-122B-A10B, Inference Model=Qwen3.5-122B-A10B2026.05 | 46.8 | 33.1 | |
| Skill-extractSkill Generation=Qwen3.5-122B-A10B, Inference Model=Qwen3.5-122B-A10B2026.05 | 45.1 | 30.9 | |
| ICLSkill Generation=None, Inference Model=Qwen3.5-122B-A10B2026.05 | 42.7 | 26.6 | |
| ReActSkill Generation=None, Inference Model=Qwen3.5-122B-A10B2026.05 | 36.7 | 23 |