Multi Turn Miss Param

BFCL · 50 models

Rank	Model	Score
1	xLAM-2-70b-fc-r (FC)	74
2	BitAgent-Bounty-8B	68
3	xLAM-2-32b-fc-r (FC)	67.5
4	xLAM-2-8b-fc-r (FC)	65
5	GLM-4.6 (FC thinking)	63
6	Claude-Opus-4-5-20251101 (FC)	58
7	xLAM-2-3b-fc-r (FC)	57.5
8	Gemini-3-Pro-Preview (FC)	56.5
9	Gemini-3-Pro-Preview (Prompt)	54.5
9	o3-2025-04-16 (Prompt)	54.5
11	Arch-Agent-32B	53
12	Claude-Haiku-4-5-20251001 (FC)	52.5
12	Claude-Sonnet-4-5-20250929 (FC)	52.5
14	Command A Reasoning (FC)	49.5
15	Nanbeige4-3B-Thinking-2511 (FC)	45
16	Moonshotai-Kimi-K2-Instruct (FC)	44.5
17	Grok-4-1-fast-reasoning (FC)	43
18	o4-mini-2025-04-16 (FC)	40.5
19	Qwen3-32B (FC)	40
19	Qwen3-8B (FC)	40
21	Grok-4-1-fast-non-reasoning (FC)	37.5
22	xLAM-2-1b-fc-r (FC)	37
23	Qwen3-32B (Prompt)	36.5
24	Grok-4-0709 (Prompt)	36
25	DeepSeek-V3.2-Exp (FC)	33.5
25	GPT-5.2-2025-12-11 (Prompt)	33.5
25	Qwen3-14B (FC)	33.5
25	Qwen3-235B-A22B-Instruct-2507 (FC)	33.5
29	GPT-4.1-2025-04-14 (FC)	32.5
29	GPT-5-nano-2025-08-07 (FC)	32.5
31	Command A (FC)	32
31	Gemini-2.5-Flash (FC)	32
33	Qwen3-235B-A22B-Instruct-2507 (Prompt)	31.5
34	Arch-Agent-3B	31
34	Qwen3-14B (Prompt)	31
36	GPT-4.1-mini-2025-04-14 (FC)	30.5
36	ToolACE-2-8B (FC)	30.5
38	Nanbeige3.5-Pro-Thinking (FC)	29
39	Grok-4-0709 (FC)	28.5
40	GPT-5.2-2025-12-11 (FC)	27.5
41	DeepSeek-V3.2-Exp (Prompt + Thinking)	27
41	Qwen3-8B (Prompt)	27
43	Qwen3-30B-A3B-Instruct-2507 (FC)	25
44	GPT-5-mini-2025-08-07 (FC)	23.5
45	Arch-Agent-1.5B	21.5
45	Claude-Opus-4-5-20251101 (Prompt)	21.5
45	Hammer2.1-7b (FC)	21.5
48	GPT-4.1-nano-2025-04-14 (FC)	17.5
48	Gemini-2.5-Flash (Prompt)	17.5
48	o4-mini-2025-04-16 (Prompt)	17.5