Latency Standard Deviation (s)

BFCL · 50 models

Rank	Model	Score
1	Amazon-Nova-Micro-v1:0 (FC)	0.45
2	Claude-Sonnet-4-5-20250929 (Prompt)	1.53
3	Llama-3.1-Nemotron-Ultra-253B-v1 (FC)	1.84
4	Amazon-Nova-Pro-v1:0 (FC)	1.91
5	xLAM-2-1b-fc-r (FC)	2.35
6	Qwen3-235B-A22B-Instruct-2507 (FC)	2.44
7	Qwen3-235B-A22B-Instruct-2507 (Prompt)	2.75
8	Command R7B (FC)	2.87
9	GPT-4.1-2025-04-14 (FC)	3.05
10	Claude-Opus-4-5-20251101 (FC)	3.13
11	Hammer2.1-0.5b (FC)	3.17
12	GPT-4.1-2025-04-14 (Prompt)	3.23
13	Granite-20b-FunctionCalling (FC)	3.43
14	Mistral-Medium-2505	3.5
15	xLAM-2-3b-fc-r (FC)	3.59
16	GPT-4.1-mini-2025-04-14 (FC)	3.65
17	Claude-Haiku-4-5-20251001 (FC)	3.92
18	Arch-Agent-1.5B	4.01
19	mistral-large-2411 (FC)	4.02
20	Qwen3-0.6B (Prompt)	4.32
21	Claude-Sonnet-4-5-20250929 (FC)	4.43
22	Gemini-2.5-Flash (Prompt)	4.44
22	Mistral-Medium-2505 (FC)	4.44
24	GPT-4.1-mini-2025-04-14 (Prompt)	4.5
25	Gemini-2.5-Flash-Lite (Prompt)	4.75
26	Granite-4.0-350m (FC)	4.85
27	Command A Reasoning (FC)	4.91
28	o4-mini-2025-04-16 (Prompt)	5.19
29	Arch-Agent-3B	6.65
30	Mistral-Small-2506 (Prompt)	6.79
31	o3-2025-04-16 (Prompt)	7.01
32	Open-Mistral-Nemo-2407 (Prompt)	7.05
33	mistral-large-2411 (Prompt)	7.15
34	o4-mini-2025-04-16 (FC)	7.18
35	GLM-4.6 (FC thinking)	7.22
36	GPT-4.1-nano-2025-04-14 (Prompt)	7.3
37	Grok-4-1-fast-non-reasoning (FC)	7.31
38	Command A (FC)	7.36
39	Qwen3-1.7B (FC)	7.37
40	palmyra-x-004 (FC)	7.62
41	GPT-5-nano-2025-08-07 (Prompt)	7.68
42	Qwen3-30B-A3B-Instruct-2507 (Prompt)	7.9
43	Gemini-2.5-Flash-Lite (FC)	8.06
44	xLAM-2-32b-fc-r (FC)	8.21
45	Qwen3-0.6B (FC)	8.45
46	Qwen3-14B (Prompt)	8.5
47	o3-2025-04-16 (FC)	8.69
48	Hammer2.1-1.5b (FC)	8.79
49	Gemini-2.5-Flash (FC)	9.22
50	Moonshotai-Kimi-K2-Instruct (FC)	9.38