List your company
ResearchFormerly WizWand

Function Calling on BFCL v4

68.8ScoreClaude-Sonnet-4.5
32.81642.15851.560.842Dec 4, 2025Dec 31, 2025Jan 28, 2026Feb 25, 2026Mar 24, 2026Apr 21, 2026May 19, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Claude-Sonnet-4.5Model Type=Proprietary2025.1268.8——
Gemma-4-31B-it-NVFP4Backbone=Gemma-4-31B-it, Quantization Method=NVFP4 W4A42026.0568.69——
Gemma-4-31B-itBackbone=Gemma-4-31B-it, Quantization Method=BF162026.0568.6——
Mix-QuantBackbone=Gemma-4-31B-it, Quantization Method=Mix-Quant2026.0568.19——
DeepSeek-V3.1-Nex-N1Model Type=Open Source, Parameter Scale=> 100B, Function Call (FC) format=true, Search API=Google Search2025.1265.3——
GLM-4.6Model Type=Open Source, Parameter Scale=> 100B2025.1262.1——
GPT-5Model Type=Proprietary2025.1261.6——
Qwen3-32B-Nex-N1Model Type=Open Source, Parameter Scale=< 100B, Function Call (FC) format=true, Search API=Google Search2025.1260.5——
Kimi-K2-thinkingModel Type=Open Source, Parameter Scale=> 100B2025.1260.4——
Qwen3.5-9BBackbone=Qwen3.5-9B, Quantization Method=BF162026.0558.96——
Mix-QuantBackbone=Qwen3.5-9B, Quantization Method=Mix-Quant2026.0557.89——
Qwen3.5-9B-NVFP4Backbone=Qwen3.5-9B, Quantization Method=NVFP4 W4A42026.0556.86——
DeepSeek-V3.1Model Type=Open Source, Parameter Scale=> 100B2025.1256.3——
Gemini-2.5-proModel Type=Proprietary2025.1254——
Gemma-4-26B-A4B-itBackbone=Gemma-4-26B-A4B-it, Quantization Method=BF162026.0553.07——
Mix-QuantBackbone=Gemma-4-26B-A4B-it, Quantization Method=Mix-Quant2026.0551.94——
Qwen3-30B-A3B-Nex-N1Model Type=Open Source, Parameter Scale=< 100B, Function Call (FC) format=true, Search API=Google Search2025.1251.9——
Gemma-4-26B-A4B-it-NVFP4Backbone=Gemma-4-26B-A4B-it, Quantization Method=NVFP4 W4A42026.0548.13——
Qwen3-32BModel Type=Open Source, Parameter Scale=< 100B2025.1245.4——
InternLM3-8B-Nex-N1Model Type=Open Source, Parameter Scale=< 100B, Function Call (FC) format=true, Search API=Google Search2025.1244.5——
Mix-QuantBackbone=Qwen3-8B, Quantization Method=Mix-Quant2026.0540.63——
Qwen3-8BBackbone=Qwen3-8B, Quantization Method=BF162026.0540.5——
Qwen3-8B-NVFP4Backbone=Qwen3-8B, Quantization Method=NVFP4 W4A42026.0538.77——
Qwen3-30B-A3BModel Type=Open Source, Parameter Scale=< 100B2025.1237.5——
Minimax-M2Model Type=Open Source, Parameter Scale=> 100B2025.1234.2——
GRPOBackbone=Qwen2.5-14B, Fine-tuning=RLHF2026.02—46.2555
GRPOBackbone=Qwen3-14B, Fine-tuning=RLHF2026.02—44.553
GSPOBackbone=Qwen2.5-14B, Fine-tuning=RLHF2026.02—50.7556
GSPOBackbone=Qwen3-14B, Fine-tuning=RLHF2026.02—45.7555
PPOBackbone=Qwen2.5-14B, Fine-tuning=RLHF2026.02—44.7556
PPOBackbone=Qwen3-14B, Fine-tuning=RLHF2026.02—43.7552
Qwen2.5-14BBackbone=Qwen2.5-14B, Fine-tuning=None2026.02—25.535.75
Qwen3-14BBackbone=Qwen3-14B, Fine-tuning=None2026.02—40.2547
SeeUPOBackbone=Qwen2.5-14B, Fine-tuning=RLHF2026.02—55.2557
SeeUPOBackbone=Qwen3-14B, Fine-tuning=RLHF2026.02—5865