Multi Turn Acc

BFCL · 50 models

Rank	Model	Score
1	xLAM-2-70b-fc-r (FC)	77.38
2	xLAM-2-8b-fc-r (FC)	70
3	xLAM-2-32b-fc-r (FC)	69.5
4	Claude-Opus-4-5-20251101 (FC)	68.38
5	GLM-4.6 (FC thinking)	68
6	Gemini-3-Pro-Preview (FC)	63.12
7	BitAgent-Bounty-8B	62.38
8	o3-2025-04-16 (Prompt)	62.25
9	Claude-Sonnet-4-5-20250929 (FC)	61.37
10	Gemini-3-Pro-Preview (Prompt)	60.75
11	Grok-4-1-fast-reasoning (FC)	58.87
12	xLAM-2-3b-fc-r (FC)	58.38
13	Arch-Agent-32B	54.25
14	Claude-Haiku-4-5-20251001 (FC)	53.62
15	Nanbeige4-3B-Thinking-2511 (FC)	51.12
16	Moonshotai-Kimi-K2-Instruct (FC)	50.63
17	Command A Reasoning (FC)	50.12
18	Qwen3-32B (FC)	47.87
19	Grok-4-0709 (Prompt)	47
20	Grok-4-1-fast-non-reasoning (FC)	46.75
21	Qwen3-235B-A22B-Instruct-2507 (FC)	45.38
22	DeepSeek-V3.2-Exp (Prompt + Thinking)	44.88
23	Qwen3-235B-A22B-Instruct-2507 (Prompt)	44.62
24	GPT-5.2-2025-12-11 (Prompt)	43.75
25	Qwen3-32B (Prompt)	43.25
26	Qwen3-8B (FC)	41.75
26	o4-mini-2025-04-16 (FC)	41.75
28	Nanbeige3.5-Pro-Thinking (FC)	40
29	GPT-4.1-2025-04-14 (FC)	38.88
30	ToolACE-2-8B (FC)	38.38
31	DeepSeek-V3.2-Exp (FC)	37.38
32	Gemini-2.5-Flash (FC)	36.25
33	xLAM-2-1b-fc-r (FC)	36
34	Arch-Agent-3B	34.88
35	Qwen3-14B (FC)	34.75
36	GPT-5-nano-2025-08-07 (FC)	34.5
37	GPT-4.1-mini-2025-04-14 (FC)	34.13
38	Grok-4-0709 (FC)	33.88
39	Qwen3-8B (Prompt)	33.38
40	Qwen3-30B-A3B-Instruct-2507 (FC)	30
41	Command A (FC)	29.5
42	GPT-5.2-2025-12-11 (FC)	28.12
43	GPT-5-mini-2025-08-07 (FC)	27.5
44	Arch-Agent-1.5B	26.62
45	Qwen3-14B (Prompt)	26.13
46	Hammer2.1-7b (FC)	23.87
47	GPT-4.1-nano-2025-04-14 (FC)	23.62
48	Qwen3-30B-A3B-Instruct-2507 (Prompt)	23.5
49	Qwen3-4B-Instruct-2507 (FC)	22.12
50	Llama-3.3-70B-Instruct (FC)	21.5