List your company

Function Calling / Tool Use on BFCL parallel parallel-multiple Actions (Accuracy)

82.2AccuracyBase
9.19228.14647.166.054May 26, 2026

Evaluation results

MethodLinks
BaseModel=Qwen2.5-7B-Instruct, Evaluation Protocol=FULL2026.0582.2
MAIGOModel=Qwen2.5-7B-Instruct, Evaluation Protocol=FULL2026.0582.1
RLSTAModel=Qwen2.5-7B-Instruct, Evaluation Protocol=FULL2026.0581.8
GRPOModel=Qwen2.5-7B-Instruct, Evaluation Protocol=FULL2026.0581.6
SFTModel=Qwen2.5-7B-Instruct, Evaluation Protocol=FULL2026.0579.3
MAIGOModel=Qwen2.5-3B-Instruct, Evaluation Protocol=FULL2026.0577.5
RLSTAModel=Qwen2.5-3B-Instruct, Evaluation Protocol=FULL2026.0573.2
BaseModel=Qwen2.5-3B-Instruct, Evaluation Protocol=FULL2026.0572.8
GRPOModel=Qwen2.5-3B-Instruct, Evaluation Protocol=FULL2026.0570.9
SFTModel=Qwen2.5-3B-Instruct, Evaluation Protocol=FULL2026.0568.5
MAIGOModel=Qwen2.5-7B-Instruct, Evaluation Protocol=SHARDED2026.0546.8
MAIGOModel=Qwen2.5-3B-Instruct, Evaluation Protocol=SHARDED2026.0540
RLSTAModel=Qwen2.5-7B-Instruct, Evaluation Protocol=SHARDED2026.0524.5
GRPOModel=Qwen2.5-7B-Instruct, Evaluation Protocol=SHARDED2026.0521.5
SFTModel=Qwen2.5-7B-Instruct, Evaluation Protocol=SHARDED2026.0518.2
BaseModel=Qwen2.5-7B-Instruct, Evaluation Protocol=SHARDED2026.0517.8
RLSTAModel=Qwen2.5-3B-Instruct, Evaluation Protocol=SHARDED2026.0517.3
GRPOModel=Qwen2.5-3B-Instruct, Evaluation Protocol=SHARDED2026.0515.8
SFTModel=Qwen2.5-3B-Instruct, Evaluation Protocol=SHARDED2026.0512.2
BaseModel=Qwen2.5-3B-Instruct, Evaluation Protocol=SHARDED2026.0512