← BFCL leaderboards

Format Variant 2

BFCL · 40 models

RankModelScore
1Claude-Sonnet-4-5-20250929 (Prompt)84
2GPT-4.1-mini-2025-04-14 (Prompt)82.5
2Gemini-3-Pro-Preview (Prompt)82.5
2Qwen3-235B-A22B-Instruct-2507 (Prompt)82.5
5GPT-4.1-2025-04-14 (Prompt)82
5Qwen3-14B (Prompt)82
5Qwen3-30B-A3B-Instruct-2507 (Prompt)82
8Qwen3-4B-Instruct-2507 (Prompt)81.5
9mistral-large-2411 (Prompt)81
10Claude-Opus-4-5-20251101 (Prompt)80
10Gemini-2.5-Flash (Prompt)80
12Mistral-Small-2506 (Prompt)79
13Grok-4-0709 (Prompt)78.5
13Qwen3-32B (Prompt)78.5
13Qwen3-8B (Prompt)78.5
13o3-2025-04-16 (Prompt)78.5
17GPT-5.2-2025-12-11 (Prompt)78
18DeepSeek-V3.2-Exp (Prompt + Thinking)77
18Gemma-3-27b-it (Prompt)77
18Open-Mistral-Nemo-2407 (Prompt)77
21Mistral-Medium-250576.5
22o4-mini-2025-04-16 (Prompt)75.5
23GPT-5-mini-2025-08-07 (Prompt)75
24Gemma-3-12b-it (Prompt)72
24Llama-3.1-8B-Instruct (Prompt)72
24Phi-4 (Prompt)72
27GPT-5-nano-2025-08-07 (Prompt)70
28Bielik-11B-v2.3-Instruct (Prompt)68
29Claude-Haiku-4-5-20251001 (Prompt)64
30MiniCPM3-4B (Prompt)60
31Gemma-3-4b-it (Prompt)52.5
32Gemini-2.5-Flash-Lite (Prompt)51.5
33RZN-T (Prompt)49.5
34Qwen3-0.6B (Prompt)48
35GPT-4.1-nano-2025-04-14 (Prompt)47
36ToolACE-2-8B (FC)41.5
37CoALM-8B7
38CoALM-70B6.5
39Gemma-3-1b-it (Prompt)4
40Ministral-8B-Instruct-2410 (FC)0