List your company

End-to-end Tool-use on ToolBench I1 v1

56.13SoPRToolLlama-2*
27.37434.839542.30549.7705Jan 29, 2026

Evaluation results

MethodLinks
ToolLlama-2*Setting=Retrieval2026.0156.1350.92
ToolGen*2026.0156.1350.92
ToolWeaverSetting=Direct Generation2026.0154.8536.08
ToolLlama*Setting=Retrieval2026.0154.649.08
ToolWeaverSetting=Direct Generation2026.0153.1740.49
ToolWeaver2026.0153.1740.49
GPT-4o-miniSetting=Retrieval (Re.)2026.0153.16—
ToolGenSetting=Direct Generation2026.0152.9736.2
ToolGen2026.0152.9736.2
GPT-4o-miniSetting=Retrieval (Re.)2026.0152.25—
GPT-4o-miniSetting=Retrieval2026.0152.2547.24
GPT-3.5*Setting=Retrieval2026.0151.4353.37
ToolGenSetting=Direct Generation2026.0145.3632.91
ToolLlama-2Setting=Retrieval (Re.)2026.0128.9425.15
ToolLlama-2Setting=Retrieval2026.0128.9425.15
ToolLlama-2Setting=Retrieval (Re.)2026.0128.4826.58