Non-Live

BFCL · 50 models

Rank	Model	Score
1	Gemini-3-Pro-Preview (Prompt)	90.65
2	Qwen3-235B-A22B-Instruct-2507 (Prompt)	90.33
3	Qwen3-32B (Prompt)	90.27
4	Mistral-Small-2506 (Prompt)	89.69
5	Claude-Opus-4-5-20251101 (Prompt)	89.65
6	xLAM-2-32b-fc-r (FC)	89.6
7	Qwen3-14B (Prompt)	89.46
8	Llama-4-Scout-17B-16E-Instruct (FC)	89.38
9	Arch-Agent-32B	88.92
10	Qwen3-30B-A3B-Instruct-2507 (Prompt)	88.92
11	Qwen3-32B (FC)	88.77
12	GPT-4.1-2025-04-14 (Prompt)	88.69
13	Llama-4-Maverick-17B-128E-Instruct-FP8 (FC)	88.65
14	Claude-Sonnet-4-5-20250929 (FC)	88.65
15	Claude-Opus-4-5-20251101 (FC)	88.58
16	Qwen3-8B (Prompt)	88.56
17	Open-Mistral-Nemo-2407 (Prompt)	88.46
18	xLAM-2-70b-fc-r (FC)	88.44
19	Grok-4-1-fast-reasoning (FC)	88.27
20	Grok-4-1-fast-non-reasoning (FC)	88.13
21	Gemini-2.5-Flash (Prompt)	88.08
22	Llama-3.3-70B-Instruct (FC)	88.02
23	Qwen3-4B-Instruct-2507 (FC)	87.88
24	Qwen3-8B (FC)	87.58
25	GLM-4.6 (FC thinking)	87.56
26	Command A (FC)	87.56
27	palmyra-x-004 (FC)	87.46
28	Gemma-3-27b-it (Prompt)	87.17
29	ToolACE-2-8B (FC)	87.1
30	Amazon-Nova-2-Lite-v1:0 (FC)	86.96
31	Arch-Agent-3B	86.67
32	Gemini-2.5-Flash-Lite (FC)	86.6
33	Amazon-Nova-Pro-v1:0 (FC)	86.58
34	Claude-Haiku-4-5-20251001 (FC)	86.5
35	Qwen3-4B-Instruct-2507 (Prompt)	86.44
36	Command A Reasoning (FC)	86.27
37	Qwen3-30B-A3B-Instruct-2507 (FC)	85.77
38	Gemini-3-Pro-Preview (FC)	85.75
39	DeepSeek-V3.2-Exp (Prompt + Thinking)	85.52
40	Hammer2.1-7b (FC)	85.5
41	Grok-4-0709 (FC)	85.38
42	Mistral-Medium-2505	85.33
43	Falcon3-10B-Instruct (FC)	85
44	Hammer2.1-3b (FC)	84.96
45	Gemini-2.5-Flash (FC)	84.96
46	Qwen3-14B (FC)	84.94
47	CoALM-8B	84.87
48	mistral-large-2411 (FC)	84.65
49	GPT-4.1-mini-2025-04-14 (Prompt)	84.6
50	xLAM-2-8b-fc-r (FC)	84.58