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

Function Calling on BFCL V3 (test)

63.34Overall Accuracy

CoALM 405B

37.350444.097750.84557.5923Feb 12, 2025
Updated 6d ago

Evaluation Results

MethodLinks
2025.02
63.3490.4684.7574.5928.2510072.26-------
2025.02
60.4982.9481.3672.1926.2572.2285.36-------
2025.02
59.8370.0860.7976.4134.6251.2287.34-------
2025.02
59.486.5285.0573.261978.0576.97-------
56.3889.7184.770.7711.7588.8970.86-------
2025.02
54.1185.1778.6172.59777.7883-------
2025.02
52.5582.1986.9871.080.8870.7387.29-------
52.1386.9483.6671.170.3895.1273.2-------
51.3684.8590.0562.517.2595.1248.33-------
49.8484.2579.7560.3310.2575.6147.92-------
2025.02
38.3556.3363.7757.310.2577.7841.84-------
2026.05
-------5276.9184.3472.5461,142.31,016.862,159.1
2026.05
-------6676.8988.3770.2673,514.61,150.374,664.9
2026.05
-------69.3981.4386.777.7568,813.41,052.369,865.7
2026.05
-------7885.1190.481.3668,012.21,03869,050.2