Share your thoughts, 1 month free Claude Pro on usSee more

Function Calling on ACEBench

78Atom Score

R2IF

Updated 1mo ago

Evaluation Results

Method	Links
R2IF 2026.04		78	57.5	96	77.17
Binary Reward 2026.04		76.7	49	96	73.9
ToolRL 2026.04		76.3	56.5	96	76.27
R2IF 2026.04		73.7	51	94	72.9
Binary Reward 2026.04		72	47.5	90	69.83
SFT 2026.04		71.3	54.5	94	73.27
R2IF 2026.04		70.7	26	90	62.23
SFT 2026.04		67.3	34.5	54	51.93
ToolRL 2026.04		65.7	39	88	64.23
R2IF 2026.04		64.7	20.5	88	57.73
SFT 2026.04		63.7	29.5	44	45.73
SFT 2026.04		61.3	23	16	33.43
ToolRL 2026.04		58	31.5	96	61.83
Raw 2026.04		57.3	40	88	61.77
ToolRL 2026.04		52.3	26	88	55.43
Raw 2026.04		52	25	90	55.67
Binary Reward 2026.04		49	20.5	60	43.17
Raw 2026.04		45.3	17	14	25.43
Raw 2026.04		43.7	13.5	30	29.07
Binary Reward 2026.04		43.3	14.5	78	45.27