← BFCL leaderboards

Latency Mean (s)

BFCL · 50 models

RankModelScore
1Qwen3-0.6B (FC)0.68
2Open-Mistral-Nemo-2407 (Prompt)0.84
3Mistral-Small-2506 (Prompt)0.92
4Gemini-2.5-Flash-Lite (Prompt)1
5GPT-4.1-nano-2025-04-14 (Prompt)1.02
6Open-Mistral-Nemo-2407 (FC)1.07
7Amazon-Nova-Micro-v1:0 (FC)1.12
8Gemini-2.5-Flash-Lite (FC)1.18
9GPT-4.1-2025-04-14 (Prompt)1.2
9Qwen3-14B (Prompt)1.2
11Mistral-Medium-25051.21
12Qwen3-30B-A3B-Instruct-2507 (Prompt)1.24
13GPT-4.1-mini-2025-04-14 (FC)1.32
14GPT-4.1-mini-2025-04-14 (Prompt)1.36
15Command R7B (FC)1.38
16Llama-3.1-Nemotron-Ultra-253B-v1 (FC)1.42
17GPT-4.1-nano-2025-04-14 (FC)1.44
18Mistral-small-2506 (FC)1.48
19Mistral-Medium-2505 (FC)1.6
20GPT-4.1-2025-04-14 (FC)1.63
21Claude-Haiku-4-5-20251001 (FC)1.68
22Granite-4.0-350m (FC)1.74
23mistral-large-2411 (Prompt)1.82
24mistral-large-2411 (FC)2.04
25Command A (FC)2.09
26GPT-5.2-2025-12-11 (FC)2.23
27Amazon-Nova-Pro-v1:0 (FC)2.25
28Grok-4-1-fast-non-reasoning (FC)2.29
29Arch-Agent-1.5B2.38
30Qwen3-235B-A22B-Instruct-2507 (Prompt)2.56
31Qwen3-235B-A22B-Instruct-2507 (FC)2.57
32Hammer2.1-0.5b (FC)2.79
33xLAM-2-1b-fc-r (FC)2.84
34Gemini-2.5-Flash (FC)2.99
35Qwen3-0.6B (Prompt)3.1
36Gemini-2.5-Flash (Prompt)3.18
37Granite-20b-FunctionCalling (FC)3.2
38Llama-3.2-1B-Instruct (FC)3.21
39Command A Reasoning (FC)3.44
40o3-2025-04-16 (FC)3.5
41Arch-Agent-3B3.56
42o4-mini-2025-04-16 (FC)3.71
42palmyra-x-004 (FC)3.71
44Claude-Haiku-4-5-20251001 (Prompt)3.75
45Claude-Opus-4-5-20251101 (Prompt)3.76
46xLAM-2-3b-fc-r (FC)3.8
47Claude-Sonnet-4-5-20250929 (Prompt)3.84
48Gemma-3-1b-it (Prompt)3.98
49GPT-5.2-2025-12-11 (Prompt)4.21
50Claude-Sonnet-4-5-20250929 (FC)4.31