List your company

Tool-use reasoning on ToolAlpaca

66.73AccuracyMOPD
44.99450.63756.2861.923May 12, 2026May 14, 2026May 16, 2026May 18, 2026May 20, 2026May 22, 2026May 25, 2026

Evaluation results

MethodLinks
MOPDbackbone=Qwen3-8B2026.0566.73
SDPObackbone=Qwen3-8B2026.0563.45
MCPO-GRPOBackbone=Qwen3-8B2026.0561.45
MCPO-DAPOBackbone=Qwen3-8B2026.0561.14
GRPObackbone=Qwen3-8B2026.0560.32
MCPO-GRPOBackbone=Qwen3-4B2026.0560.29
MCPO-DAPOBackbone=Qwen3-4B2026.0559.87
MT-GRPOBackbone=Qwen3-8B2026.0559.4
Qwen3-8B2026.0559.11
MGSBackbone=Qwen3-8B2026.0558.82
MT-GRPOBackbone=Qwen3-4B2026.0558.67
MGSBackbone=Qwen3-4B2026.0558.35
CLIPOBackbone=Qwen3-8B2026.0557.69
GRPOBackbone=Qwen3-8B2026.0557.32
GRPOBackbone=Qwen3-4B2026.0556.73
DAPOBackbone=Qwen3-8B2026.0556.42
CLIPOBackbone=Qwen3-4B2026.0555.56
DAPOBackbone=Qwen3-4B2026.0555.29
Base ModelBackbone=Qwen3-8B2026.0547.35
Base ModelBackbone=Qwen3-4B2026.0545.83