List your company

Tool Use on tool-use (test)

72AccuracyPBSD
35.49644.97354.4563.927May 6, 2026

Evaluation results

MethodLinks
PBSDModel=Qwen3-8B2026.0572
GRPOModel=Qwen3-8B2026.0568.8
DAPOModel=Qwen3-8B2026.0567.7
OPSDModel=Qwen3-8B2026.0565.6
SDFTModel=Qwen3-8B2026.0562.8
SRPOModel=Qwen3-8B2026.0562.1
SFTModel=Qwen3-8B2026.0561.7
Base (Instruct)Model=Qwen3-8B2026.0561.3
PBSDModel=Qwen3-4B2026.0560.6
GRPOModel=Qwen3-4B2026.0558.9
OPSDModel=Qwen3-4B2026.0553.9
SFTModel=Qwen3-4B2026.0551.1
DAPOModel=Qwen3-4B2026.0548.2
SRPOModel=Qwen3-4B2026.0546.1
Base (Instruct)Model=Qwen3-4B2026.0545.7
PBSDModel=Qwen3-1.7B2026.0544.7
DAPOModel=Qwen3-1.7B2026.0543.6
GRPOModel=Qwen3-1.7B2026.0542.9
SDFTModel=Qwen3-4B2026.0541.8
SFTModel=Qwen3-1.7B2026.0539.7
SDFTModel=Qwen3-1.7B2026.0538.3
OPSDModel=Qwen3-1.7B2026.0537.2
Base (Instruct)Model=Qwen3-1.7B2026.0536.9
SRPOModel=Qwen3-1.7B2026.0536.9