List your company

Function Calling on BFCL held-out V4 (test)

52.4Average ScoreCURATEEVO
10.90421.67732.4543.223Jul 7, 2026

Evaluation results

MethodLinks
CURATEEVOModel=QWEN3-4B, Data Setting=Labeled Data2026.0752.4
FunReason-MTModel=QWEN3-4B, Data Setting=Labeled Data2026.0750.3
CURATEEVOModel=QWEN3-4B, Data Setting=Wild Data2026.0750
FunReason-MTModel=QWEN3-4B, Data Setting=Wild Data2026.0748.7
EnvScalerModel=QWEN3-8B, Data Setting=Labeled Data2026.0747.6
RODSModel=QWEN3-4B, Data Setting=Labeled Data2026.0747.2
EnvScalerModel=QWEN3-8B, Data Setting=Wild Data2026.0746.2
RODSModel=QWEN3-4B, Data Setting=Wild Data2026.0745.9
MUA-RLModel=QWEN3-8B, Data Setting=Labeled Data2026.0742.6
MUA-RLModel=QWEN3-8B, Data Setting=Wild Data2026.0740.9
AWMModel=QWEN3-8B, Data Setting=Labeled Data2026.0739.7
AWMModel=QWEN3-8B, Data Setting=Wild Data2026.0738.4
GRPO w/o. Data CurationModel=QWEN3-8B, Data Setting=Labeled Data2026.0718.7
GRPO w/o. Data CurationModel=QWEN3-8B, Data Setting=Wild Data2026.0717.4
GRPO w/o. Data CurationModel=QWEN3-4B, Data Setting=Labeled Data2026.0713.8
GRPO w/o. Data CurationModel=QWEN3-4B, Data Setting=Wild Data2026.0712.5