List your company

Tool Use on ToolHop

54.87Answer CorrectnessHarnessForge
28.204435.127242.0548.9728Jun 1, 2026

Evaluation results

MethodLinks
HarnessForgeBackbone=Qwen3-8B2026.0654.8774.05
HarnessForgeBackbone=Qwen3-4B2026.0652.8268.1
GRPOBackbone=Qwen3-8B2026.0651.2875.62
RLOOBackbone=Qwen3-8B2026.0650.7772.75
GRPOBackbone=Qwen3-4B2026.0649.7466.41
SFTBackbone=Qwen3-8B2026.0648.7272.2
MermaidFlowBackbone=Qwen3-8B2026.0647.3959.31
RLOOBackbone=Qwen3-4B2026.0646.1564.87
SFTBackbone=Qwen3-4B2026.0645.1363.9
MermaidFlowBackbone=Qwen3-4B2026.0644.6258.74
MaASBackbone=Qwen3-8B2026.0644.6251.8
MaASBackbone=Qwen3-4B2026.0642.0551.8
ADASBackbone=Qwen3-8B2026.0642.0561.32
ADASBackbone=Qwen3-4B2026.064044.99
AFlowBackbone=Qwen3-8B2026.0632.8254.12
AFlowBackbone=Qwen3-4B2026.0631.2843.81
AgentSquareBackbone=Qwen3-8B2026.0630.7741.22
AgentSquareBackbone=Qwen3-4B2026.0629.2339.46