List your company

Tool Planning on ToolBench (EM, ELR, ACPL)

42.6EM (%)GRAFT
4.95214.72624.534.274May 12, 2026

Evaluation results

MethodLinks
GRAFTLLM=Mistral-7B2026.0542.6401.35
GRAFTLLM=Llama-3.2-3B2026.0541.8361.33
GRAFTLLM=Qwen2.5-1.5B2026.0539.4351.28
ToolGenLLM=Mistral-7B2026.0535331.21
ToolGenLLM=Llama-3.2-3B2026.0533311.18
ToolGenLLM=Qwen2.5-1.5B2026.0531.4281.03
PLaGLLM=Mistral-7B2026.0521.95390.9
PLaGLLM=Llama-3.2-3B2026.0516.86330.67
HuggingGPTLLM=Mistral-7B2026.0516.01380.6
DirectLLM=Llama-3.2-3B2026.0515.58220.57
DirectLLM=Mistral-7B2026.0515.47370.52
PLaGLLM=Qwen2.5-1.5B2026.0515.29340.65
HuggingGPTLLM=Llama-3.2-3B2026.0515.15250.57
GToolLLM=Mistral-7B2026.0513.8240.58
BeamSearchLLM=Mistral-7B2026.0513.26510.57
GToolLLM=Llama-3.2-3B2026.0511.8220.52
BeamSearchLLM=Llama-3.2-3B2026.0511.7740.57
GToolLLM=Qwen2.5-1.5B2026.0510.8210.51
BeamSearchLLM=Qwen2.5-1.5B2026.0510.59670.51
HuggingGPTLLM=Qwen2.5-1.5B2026.059.21320.36
DirectLLM=Qwen2.5-1.5B2026.059.15310.32
GNN4PlanLLM=Mistral-7B2026.056.71000.72
GNN4PlanLLM=Llama-3.2-3B2026.056.611000.7
GNN4PlanLLM=Qwen2.5-1.5B2026.056.41000.7