List your company

Tool Use on BFCL (v4, v3, v4+H)

77BFCL v4 ScoreVanilla
-0.4819.63539.7559.865Jul 5, 2026

Evaluation results

MethodLinks
VanillaModel=Opus 4.72026.07774285.5
NREModel=Haiku 4.52026.077341.591.5
VanillaModel=Sonnet 4.62026.07734583.5
NREModel=Kimi-k2.62026.0770.542.589
NREModel=Mistral 3.52026.0770.543.585
NREModel=Deepseek-v32026.0770.539.587.5
VanillaModel=Kimi-k2.62026.07703975
GEPAModel=Kimi-k2.62026.07704187.5
NREModel=gpt-5.4-mini2026.07674485.5
NREModel=gpt-oss-120b2026.07643690
VanillaModel=Haiku 4.52026.07624273
VanillaModel=gpt-oss-120b2026.07613069.5
GEPAModel=gpt-oss-120b2026.0757.53174
VanillaModel=Deepseek-R12026.075730.575.5
RLMModel=Kimi-k2.62026.07562968
VanillaModel=Deepseek-v32026.0746.52861.5
VanillaModel=gpt-5.4-mini2026.0738.52256
GEPAModel=Deepseek-v32026.0726.519.541.5
RLMModel=gpt-oss-120b2026.0715.5128
RLMModel=Deepseek-v32026.072.51.55.5