List your company

Tool-use agentic performance on AppWorld

64.88Avg@4SkillX
25.266435.550745.83556.1193Apr 6, 2026

Evaluation results

MethodLinks
SkillXModel=GLM-4.6, Experience extraction model alignment=Aligned with inference model2026.0464.8888.69
ExpeLModel=GLM-4.6, Experience extraction model alignment=Aligned with inference model2026.0464.1485.12
AWMModel=GLM-4.6, Experience extraction model alignment=Aligned with inference model2026.0462.284.52
A-MemModel=GLM-4.6, Experience extraction model alignment=Aligned with inference model2026.0460.5783.93
No MemoryModel=GLM-4.6, Experience extraction model alignment=Aligned with inference model2026.0460.2783.33
SkillXModel=Kimi-K2-Instruct-0905, Experience extraction model alignment=GLM-4.6 used for experience extraction2026.0456.481.55
ExpeLModel=Kimi-K2-Instruct-0905, Experience extraction model alignment=GLM-4.6 used for experience extraction2026.0452.9878.87
ExpeLModel=Kimi-K2-Instruct-0905, Experience extraction model alignment=Aligned with inference model2026.0452.5378.57
AWMModel=Kimi-K2-Instruct-0905, Experience extraction model alignment=GLM-4.6 used for experience extraction2026.0450.676.49
AWMModel=Kimi-K2-Instruct-0905, Experience extraction model alignment=Aligned with inference model2026.0449.776.19
No MemoryModel=Kimi-K2-Instruct-0905, Experience extraction model alignment=Aligned with inference model2026.0446.8870.24
A-MemModel=Kimi-K2-Instruct-0905, Experience extraction model alignment=Aligned with inference model2026.0446.5872.62
SkillXModel=Qwen3-32B, Experience extraction model alignment=GLM-4.6 used for experience extraction2026.0435.1258.93
AWMModel=Qwen3-32B, Experience extraction model alignment=GLM-4.6 used for experience extraction2026.0434.4556.25
ExpeLModel=Qwen3-32B, Experience extraction model alignment=GLM-4.6 used for experience extraction2026.0432.9458.78
ExpeLModel=Qwen3-32B, Experience extraction model alignment=Aligned with inference model2026.0432.8758.93
AWMModel=Qwen3-32B, Experience extraction model alignment=Aligned with inference model2026.0430.855.95
No MemoryModel=Qwen3-32B, Experience extraction model alignment=Aligned with inference model2026.0427.6847.62
A-MemModel=Qwen3-32B, Experience extraction model alignment=Aligned with inference model2026.0426.7950.59