← BFCL leaderboards

Format Variant 23

BFCL · 40 models

RankModelScore
1Claude-Sonnet-4-5-20250929 (Prompt)84
2Gemini-3-Pro-Preview (Prompt)82.5
3GPT-4.1-mini-2025-04-14 (Prompt)77
4GPT-5.2-2025-12-11 (Prompt)76.5
5Claude-Opus-4-5-20251101 (Prompt)76
5Qwen3-235B-A22B-Instruct-2507 (Prompt)76
5o4-mini-2025-04-16 (Prompt)76
8Gemini-2.5-Flash (Prompt)75.5
8Grok-4-0709 (Prompt)75.5
10o3-2025-04-16 (Prompt)74.5
11Mistral-Small-2506 (Prompt)73.5
12DeepSeek-V3.2-Exp (Prompt + Thinking)73
12GPT-5-mini-2025-08-07 (Prompt)73
12Qwen3-14B (Prompt)73
15Qwen3-30B-A3B-Instruct-2507 (Prompt)72.5
16mistral-large-2411 (Prompt)72
17Gemma-3-27b-it (Prompt)71
17Mistral-Medium-250571
19Qwen3-32B (Prompt)70
20GPT-5-nano-2025-08-07 (Prompt)69.5
20Open-Mistral-Nemo-2407 (Prompt)69.5
20Qwen3-8B (Prompt)69.5
23Gemini-2.5-Flash-Lite (Prompt)68.5
23Phi-4 (Prompt)68.5
25Qwen3-4B-Instruct-2507 (Prompt)68
26GPT-4.1-2025-04-14 (Prompt)63.5
27Llama-3.1-8B-Instruct (Prompt)62.5
28Bielik-11B-v2.3-Instruct (Prompt)55.5
29GPT-4.1-nano-2025-04-14 (Prompt)47.5
30MiniCPM3-4B (Prompt)45
31Gemma-3-12b-it (Prompt)42.5
32Gemma-3-4b-it (Prompt)29
33Claude-Haiku-4-5-20251001 (Prompt)25.5
34Qwen3-0.6B (Prompt)15
35ToolACE-2-8B (FC)13
36RZN-T (Prompt)11
37CoALM-70B0
37CoALM-8B0
37Gemma-3-1b-it (Prompt)0
37Ministral-8B-Instruct-2410 (FC)0