Irrelevance Detection

BFCL · 50 models

Rank	Model	Score
1	Llama-3.1-Nemotron-Ultra-253B-v1 (FC)	100
1	Ministral-8B-Instruct-2410 (FC)	100
3	ToolACE-2-8B (FC)	97.08
4	BitAgent-Bounty-8B	96.67
5	Falcon3-1B-Instruct (FC)	95.42
6	Gemini-2.5-Flash (FC)	94.58
6	Gemini-2.5-Flash-Lite (Prompt)	94.58
8	DeepSeek-V3.2-Exp (FC)	94.17
8	Gemini-2.5-Flash (Prompt)	94.17
8	Gemini-2.5-Flash-Lite (FC)	94.17
11	Claude-Haiku-4-5-20251001 (Prompt)	93.75
11	Mistral-Medium-2505 (FC)	93.75
13	Claude-Sonnet-4-5-20250929 (Prompt)	93.33
14	Claude-Opus-4-5-20251101 (Prompt)	92.92
15	GPT-5.2-2025-12-11 (Prompt)	92.5
16	GPT-5-mini-2025-08-07 (FC)	92.08
17	CoALM-70B	91.67
17	CoALM-8B	91.67
17	GPT-5-nano-2025-08-07 (FC)	91.67
17	Hammer2.1-7b (FC)	91.67
21	Hammer2.1-3b (FC)	91.25
22	Qwen3-14B (Prompt)	90.42
23	GPT-4.1-2025-04-14 (FC)	90
24	GPT-4.1-2025-04-14 (Prompt)	89.58
24	o4-mini-2025-04-16 (Prompt)	89.58
26	Gemini-3-Pro-Preview (Prompt)	89.17
26	Granite-20b-FunctionCalling (FC)	89.17
28	Qwen3-4B-Instruct-2507 (FC)	88.75
29	Mistral-small-2506 (FC)	88.33
29	Phi-4 (Prompt)	88.33
29	o3-2025-04-16 (FC)	88.33
32	Claude-Sonnet-4-5-20250929 (FC)	87.92
32	GPT-4.1-nano-2025-04-14 (Prompt)	87.92
32	Moonshotai-Kimi-K2-Instruct (FC)	87.92
35	Grok-4-0709 (Prompt)	87.5
35	Qwen3-8B (Prompt)	87.5
37	Command A Reasoning (FC)	87.08
37	Granite-3.2-8B-Instruct (FC)	87.08
37	Hammer2.1-1.5b (FC)	87.08
37	Qwen3-235B-A22B-Instruct-2507 (Prompt)	87.08
37	o3-2025-04-16 (Prompt)	87.08
42	Granite-3.1-8B-Instruct (FC)	86.67
43	Command A (FC)	86.25
43	o4-mini-2025-04-16 (FC)	86.25
45	Amazon-Nova-2-Lite-v1:0 (FC)	85.83
45	Claude-Opus-4-5-20251101 (FC)	85.83
45	Qwen3-14B (FC)	85.83
48	GLM-4.6 (FC thinking)	85.42
49	Arch-Agent-32B	85
49	Hammer2.1-0.5b (FC)	85