Multi Turn Long Context

BFCL · 50 models

Rank	Model	Score
1	xLAM-2-70b-fc-r (FC)	76
2	Claude-Opus-4-5-20251101 (FC)	70.5
3	xLAM-2-8b-fc-r (FC)	67
4	GLM-4.6 (FC thinking)	66.5
5	Gemini-3-Pro-Preview (FC)	64
5	Gemini-3-Pro-Preview (Prompt)	64
7	o3-2025-04-16 (Prompt)	63
8	Grok-4-1-fast-reasoning (FC)	62.5
9	Claude-Sonnet-4-5-20250929 (FC)	59
10	BitAgent-Bounty-8B	57
11	xLAM-2-32b-fc-r (FC)	56.5
12	Claude-Haiku-4-5-20251001 (FC)	56
13	Qwen3-235B-A22B-Instruct-2507 (FC)	55.5
14	Moonshotai-Kimi-K2-Instruct (FC)	55
15	Grok-4-1-fast-non-reasoning (FC)	52
16	Grok-4-0709 (Prompt)	50.5
16	Qwen3-235B-A22B-Instruct-2507 (Prompt)	50.5
18	Command A Reasoning (FC)	48.5
18	DeepSeek-V3.2-Exp (Prompt + Thinking)	48.5
20	Nanbeige4-3B-Thinking-2511 (FC)	47
21	GPT-5.2-2025-12-11 (Prompt)	46.5
22	ToolACE-2-8B (FC)	46
23	o4-mini-2025-04-16 (FC)	45.5
23	xLAM-2-3b-fc-r (FC)	45.5
25	Grok-4-0709 (FC)	44
26	GPT-4.1-2025-04-14 (FC)	43
26	Qwen3-32B (FC)	43
28	Arch-Agent-32B	41.5
29	Nanbeige3.5-Pro-Thinking (FC)	41
29	Qwen3-30B-A3B-Instruct-2507 (FC)	41
31	GPT-4.1-mini-2025-04-14 (FC)	40
32	GPT-5-nano-2025-08-07 (FC)	38
33	Qwen3-32B (Prompt)	36.5
34	Gemini-2.5-Flash (FC)	35.5
35	DeepSeek-V3.2-Exp (FC)	35
36	Qwen3-8B (FC)	34.5
37	GPT-5-mini-2025-08-07 (FC)	33
38	Qwen3-14B (FC)	32.5
39	GPT-5.2-2025-12-11 (FC)	30.5
40	GPT-4.1-nano-2025-04-14 (FC)	30
41	Arch-Agent-3B	29
41	Qwen3-30B-A3B-Instruct-2507 (Prompt)	29
43	Llama-3.3-70B-Instruct (FC)	26.5
43	Qwen3-8B (Prompt)	26.5
45	Qwen3-4B-Instruct-2507 (FC)	25.5
45	xLAM-2-1b-fc-r (FC)	25.5
47	Command A (FC)	25
48	Arch-Agent-1.5B	22
49	Hammer2.1-7b (FC)	21.5
50	Qwen3-4B-Instruct-2507 (Prompt)	20