← BFCL leaderboards

Format Variant 3

BFCL · 40 models

RankModelScore
1Gemini-3-Pro-Preview (Prompt)83.5
1Qwen3-14B (Prompt)83.5
3Claude-Opus-4-5-20251101 (Prompt)83
4Qwen3-30B-A3B-Instruct-2507 (Prompt)82.5
4Qwen3-32B (Prompt)82.5
6Claude-Sonnet-4-5-20250929 (Prompt)82
6Qwen3-235B-A22B-Instruct-2507 (Prompt)82
8GPT-4.1-mini-2025-04-14 (Prompt)81.5
9Qwen3-4B-Instruct-2507 (Prompt)80.5
10Gemini-2.5-Flash (Prompt)80
11Phi-4 (Prompt)79
12Qwen3-8B (Prompt)78.5
12mistral-large-2411 (Prompt)78.5
14Mistral-Small-2506 (Prompt)78
15GPT-4.1-2025-04-14 (Prompt)77.5
16Gemma-3-27b-it (Prompt)77
17DeepSeek-V3.2-Exp (Prompt + Thinking)76.5
17Open-Mistral-Nemo-2407 (Prompt)76.5
19GPT-5.2-2025-12-11 (Prompt)75
20GPT-5-mini-2025-08-07 (Prompt)74
21Gemma-3-12b-it (Prompt)72.5
21o3-2025-04-16 (Prompt)72.5
23Grok-4-0709 (Prompt)70.5
24o4-mini-2025-04-16 (Prompt)70
25GPT-5-nano-2025-08-07 (Prompt)68.5
26Llama-3.1-8B-Instruct (Prompt)68
27Mistral-Medium-250567.5
28Bielik-11B-v2.3-Instruct (Prompt)65
29Claude-Haiku-4-5-20251001 (Prompt)62.5
30Gemma-3-4b-it (Prompt)53
31MiniCPM3-4B (Prompt)48.5
32Gemini-2.5-Flash-Lite (Prompt)47
33RZN-T (Prompt)35.5
33ToolACE-2-8B (FC)35.5
35Qwen3-0.6B (Prompt)34
36GPT-4.1-nano-2025-04-14 (Prompt)32
37CoALM-70B9
38Gemma-3-1b-it (Prompt)6
39CoALM-8B1
40Ministral-8B-Instruct-2410 (FC)0