Berkeley Function Calling 大模型工具使用排行榜

Name: Berkeley Function Calling 大模型工具使用排行榜
Creator: DataLearner
License: https://creativecommons.org/licenses/by/4.0/

Berkeley Function Calling Leaderboard是衡量大模型工具使用或函数调用能力的权威排行榜。

榜首模型

Claude-Opus-4-5-20251101 (FC)

最高得分

模型数量

109

数据版本

20260705

数据来源: Berkeley官方网站

来源：全部国产模型

榜单历史快照月份:

排名总表

排名	模型名称	综合准确率	请求成本($)	时延(秒)	AST准确率(Non-Live)	实时任务准确率(Live)	相关性检测	发布者	开源情况
	Claude-Opus-4-5-20251101 (FC)Anthropic	77.47	86.55	4.38	88.58	79.79	62.50	Anthropic	Proprietary
	Claude-Sonnet-4-5-20250929 (FC)Anthropic	73.24	43.73	4.31	88.65	81.13	68.75	Anthropic	Proprietary
	Gemini-3-Pro-Preview (Prompt)Google	72.51	298.47	12.08	90.65	83.12	68.75	Google	Proprietary
4	GLM-4.6 (FC thinking)Zhipu AI	72.38	4.64	4.34	87.56	80.90	75.00	Zhipu AI	MIT
109	Gemma-3-1b-it (Prompt)Google	7.17	3.40	3.98	20.21	11.84	37.50	Google	gemma-terms-of-use
5	Grok-4-1-fast-reasoning (FC)xAI	69.57	17.26	6.74	88.27	78.46	81.25	xAI	Proprietary
6	Claude-Haiku-4-5-20251001 (FC)Anthropic	68.70	14.23	1.68	86.50	78.68	62.50	Anthropic	Proprietary
7	Gemini-3-Pro-Preview (FC)Google	68.14	224.69	15.87	85.75	81.72	75.00	Google	Proprietary
8	o3-2025-04-16 (Prompt)OpenAI	63.05	234.64	4.83	81.94	73.21	93.75	OpenAI	Proprietary
9	Grok-4-0709 (Prompt)xAI	62.97	348.19	30.38	82.75	72.54	81.25	xAI	Proprietary
10	Grok-4-0709 (FC)xAI	61.38	355.17	15.49	85.38	75.57	87.50	xAI	Proprietary
11	Moonshotai-Kimi-K2-Instruct (FC)MoonshotAI	59.06	6.19	6.40	81.60	78.68	75.00	MoonshotAI	modified-mit
12	Grok-4-1-fast-non-reasoning (FC)xAI	58.29	16.27	2.29	88.13	77.94	81.25	xAI	Proprietary
13	Command A Reasoning (FC)Cohere	57.06	3.04	3.44	86.27	78.61	68.75	Cohere	CC-BY-NC 4.0 License (w/ Acceptable Use Addendum)
14	DeepSeek-V3.2-Exp (Prompt + Thinking)DeepSeek	56.73	57.75	37.89	85.52	76.02	93.75	DeepSeek	MIT
15	Gemini-2.5-Flash (FC)Google	56.24	26.36	2.99	84.96	74.39	75.00	Google	Proprietary
16	GPT-5.2-2025-12-11 (FC)OpenAI	55.87	85.65	2.23	81.85	70.39	75.00	OpenAI	Proprietary
17	GPT-5-mini-2025-08-07 (FC)OpenAI	55.46	22.18	8.32	69.85	58.62	62.50	OpenAI	Proprietary
18	xLAM-2-32b-fc-r (FC)Salesforce	54.66	6.00	6.94	89.60	75.50	81.25	Salesforce	cc-by-nc-4.0
19	DeepSeek-V3.2-Exp (FC)DeepSeek	54.12	6.71	5.83	34.85	53.66	37.50	DeepSeek	MIT
20	GPT-4.1-2025-04-14 (FC)OpenAI	53.96	100.75	1.63	82.79	69.95	87.50	OpenAI	Proprietary
21	o4-mini-2025-04-16 (FC)OpenAI	53.24	81.91	3.71	37.73	66.10	81.25	OpenAI	Proprietary
22	xLAM-2-70b-fc-r (FC)Salesforce	53.07	25.10	28.06	88.44	72.17	75.00	Salesforce	cc-by-nc-4.0
23	Qwen3-235B-A22B-Instruct-2507 (Prompt)Qwen	52.15	3.12	2.56	90.33	78.68	93.75	Qwen	apache-2.0
24	GPT-5-nano-2025-08-07 (FC)OpenAI	51.45	8.79	10.36	68.00	59.44	75.00	OpenAI	Proprietary
25	Nanbeige4-3B-Thinking-2511 (FC)Nanbeige	51.40	14.14	13.46	81.58	79.42	75.00	Nanbeige	apache-2.0
26	Gemini-2.5-Flash (Prompt)Google	50.90	33.45	3.18	88.08	78.16	62.50	Google	Proprietary
27	GPT-4.1-mini-2025-04-14 (FC)OpenAI	50.45	19.25	1.32	83.83	68.84	81.25	OpenAI	Proprietary
28	o4-mini-2025-04-16 (Prompt)OpenAI	50.26	133.63	4.47	81.29	70.76	81.25	OpenAI	Proprietary
29	Qwen3-32B (FC)Qwen	48.71	153.08	169.87	88.77	82.01	93.75	Qwen	apache-2.0
30	o3-2025-04-16 (FC)OpenAI	48.56	133.45	3.50	40.38	66.17	81.25	OpenAI	Proprietary
31	Qwen3-235B-A22B-Instruct-2507 (FC)Qwen	47.99	2.50	2.57	37.40	68.91	87.50	Qwen	apache-2.0
32	Nanbeige3.5-Pro-Thinking (FC)Nanbeige	47.68	23.46	21.12	38.35	69.95	100.00	Nanbeige	apache-2.0
33	Qwen3-32B (Prompt)Qwen	46.78	199.47	167.54	90.27	82.01	81.25	Qwen	apache-2.0
34	xLAM-2-8b-fc-r (FC)Salesforce	46.68	20.92	22.65	84.58	67.95	87.50	Salesforce	cc-by-nc-4.0
35	Command A (FC)Cohere	46.49	91.37	2.09	87.56	78.53	81.25	Cohere	CC-BY-NC 4.0 License (w/ Acceptable Use Addendum)
36	BitAgent-Bounty-8BBittensor	46.23	18.02	16.52	81.60	93.12	68.75	Bittensor	Apache-2.0
37	Arch-Agent-32Bkatanemo	45.37	8.87	9.44	88.92	80.68	81.25	katanemo	katanemo-research
38	GPT-5.2-2025-12-11 (Prompt)OpenAI	45.27	164.58	4.21	78.29	67.14	75.00	OpenAI	Proprietary
39	Qwen3-8B (FC)Qwen	42.57	43.32	51.36	87.58	80.53	93.75	Qwen	apache-2.0
40	ToolACE-2-8B (FC)Huawei Noah & USTC	42.44	24.43	15.95	87.10	77.42	75.00	Huawei Noah & USTC	Apache-2.0
41	Qwen3-30B-A3B-Instruct-2507 (FC)Qwen	41.39	5.62	5.95	85.77	77.94	81.25	Qwen	apache-2.0
42	xLAM-2-3b-fc-r (FC)Salesforce	41.22	3.36	3.80	82.96	62.92	87.50	Salesforce	cc-by-nc-4.0
43	Qwen3-14B (FC)Qwen	41.03	3.38	4.50	84.94	80.01	87.50	Qwen	apache-2.0
44	Qwen3-8B (Prompt)Qwen	40.43	63.95	54.17	88.56	80.09	75.00	Qwen	apache-2.0
45	GPT-4.1-2025-04-14 (Prompt)OpenAI	39.38	145.85	1.20	88.69	78.90	100.00	OpenAI	Proprietary
46	mistral-large-2411 (FC)Mistral AI	38.37	115.98	2.04	84.65	81.87	93.75	Mistral AI	Proprietary
47	Qwen3-14B (Prompt)Qwen	37.77	1.35	1.20	89.46	79.35	81.25	Qwen	apache-2.0
48	Mistral-Medium-2505Mistral AI	37.69	36.51	1.21	85.33	66.03	75.00	Mistral AI	Proprietary
49	Mistral-Medium-2505 (FC)Mistral AI	37.56	18.80	1.60	67.44	67.95	62.50	Mistral AI	Proprietary
50	Llama-4-Maverick-17B-128E-Instruct-FP8 (FC)Meta	37.29	18.25	18.43	88.65	73.65	100.00	Meta	Meta Llama 4 Community
51	Mistral-small-2506 (FC)Mistral AI	37.15	5.20	1.48	73.60	77.28	87.50	Mistral AI	Proprietary
52	Gemini-2.5-Flash-Lite (FC)Google	36.87	7.55	1.18	86.60	65.80	43.75	Google	Proprietary
53	Qwen3-30B-A3B-Instruct-2507 (Prompt)Qwen	36.70	1.56	1.24	88.92	78.39	93.75	Qwen	apache-2.0
54	Qwen3-4B-Instruct-2507 (FC)Qwen	35.68	6.37	7.61	87.88	76.39	87.50	Qwen	apache-2.0
55	Qwen3-4B-Instruct-2507 (Prompt)Qwen	35.52	53.66	44.70	86.44	74.69	87.50	Qwen	apache-2.0
56	Arch-Agent-3Bkatanemo	35.36	3.70	3.56	86.67	72.91	68.75	katanemo	katanemo-research
57	Claude-Opus-4-5-20251101 (Prompt)Anthropic	33.47	88.33	3.76	89.65	76.02	68.75	Anthropic	Proprietary
58	GPT-4.1-nano-2025-04-14 (FC)OpenAI	33.05	5.66	1.44	72.98	60.77	93.75	OpenAI	Proprietary
59	Mistral-Small-2506 (Prompt)Mistral AI	32.38	6.91	0.92	89.69	79.05	93.75	Mistral AI	Proprietary
60	Arch-Agent-1.5Bkatanemo	32.14	2.45	2.38	82.67	67.73	75.00	katanemo	katanemo-research
61	Command R7B (FC)Cohere	32.07	1.50	1.38	80.96	69.06	68.75	Cohere	cc-by-nc-4.0
62	Llama-3.3-70B-Instruct (FC)Meta	31.90	29.54	26.11	88.02	76.61	100.00	Meta	Meta Llama 3 Community
63	mistral-large-2411 (Prompt)Mistral AI	31.84	232.42	1.82	83.00	68.10	93.75	Mistral AI	Proprietary
64	Hammer2.1-7b (FC)MadeAgents	31.67	4.99	5.77	85.50	69.50	50.00	MadeAgents	cc-by-nc-4.0
65	xLAM-2-1b-fc-r (FC)Salesforce	30.44	2.79	2.84	69.04	55.14	87.50	Salesforce	cc-by-nc-4.0
66	Gemma-3-12b-it (Prompt)Google	30.43	10.77	11.10	79.44	74.24	93.75	Google	gemma-terms-of-use
67	GPT-4.1-mini-2025-04-14 (Prompt)OpenAI	29.73	20.52	1.36	84.60	74.76	87.50	OpenAI	Proprietary
68	Hammer2.1-3b (FC)MadeAgents	29.71	10.89	11.24	84.96	70.54	56.25	MadeAgents	qwen-research
69	Gemma-3-27b-it (Prompt)Google	29.47	11.82	10.88	87.17	74.54	81.25	Google	gemma-terms-of-use
70	Phi-4 (Prompt)Microsoft	28.79	8.72	9.49	69.56	60.70	50.00	Microsoft	MIT
71	Qwen3-1.7B (FC)Qwen	28.41	4.33	5.12	82.92	74.61	81.25	Qwen	apache-2.0
72	Llama-4-Scout-17B-16E-Instruct (FC)Meta	28.13	24.68	17.86	89.38	74.69	100.00	Meta	Meta Llama 4 Community
73	Gemini-2.5-Flash-Lite (Prompt)Google	28.03	7.05	1.00	83.90	54.85	50.00	Google	Proprietary
74	CoALM-70BUIUC + Oumi	27.99	19.89	16.22	83.44	67.28	93.75	UIUC + Oumi	Meta Llama 3 Community
75	Hammer2.1-1.5b (FC)MadeAgents	27.88	6.83	6.28	82.98	69.50	75.00	MadeAgents	cc-by-nc-4.0
76	palmyra-x-004 (FC)Writer	27.87	178.15	3.71	87.46	77.87	81.25	Writer	Proprietary
77	GPT-5-mini-2025-08-07 (Prompt)OpenAI	27.83	82.74	8.89	68.04	62.55	93.75	OpenAI	Proprietary
78	Open-Mistral-Nemo-2407 (FC)Mistral AI	27.63	8.12	1.07	82.81	73.80	81.25	Mistral AI	Proprietary
79	GPT-5-nano-2025-08-07 (Prompt)OpenAI	27.55	21.47	10.67	80.81	70.69	93.75	OpenAI	Proprietary
80	Amazon-Nova-2-Lite-v1:0 (FC)Amazon	27.10	78.19	8.55	86.96	80.83	75.00	Amazon	Proprietary
81	Granite-3.1-8B-Instruct (FC)IBM	27.10	9.32	13.23	78.33	60.33	68.75	IBM	Apache-2.0
82	Falcon3-10B-Instruct (FC)TII UAE	27.01	52.59	69.27	85.00	75.43	93.75	TII UAE	falcon-llm-license
83	Granite-3.2-8B-Instruct (FC)IBM	26.87	25.02	36.13	79.77	60.33	75.00	IBM	Apache-2.0
84	CoALM-8BUIUC + Oumi	26.81	25.33	20.36	84.87	66.77	87.50	UIUC + Oumi	Meta Llama 3 Community
85	Llama-3.1-8B-Instruct (Prompt)Meta	25.83	7.49	5.60	84.00	70.76	93.75	Meta	Meta Llama 3 Community
86	MiniCPM3-4B-FC (FC)openbmb	25.55	54.05	118.62	81.75	65.21	68.75	openbmb	Apache-2.0
87	Claude-Haiku-4-5-20251001 (Prompt)Anthropic	25.26	45.13	3.75	55.42	52.48	31.25	Anthropic	Proprietary
88	Amazon-Nova-Pro-v1:0 (FC)Amazon	24.97	48.44	2.25	86.58	78.53	93.75	Amazon	Proprietary
89	Claude-Sonnet-4-5-20250929 (Prompt)Anthropic	24.90	47.82	3.84	59.81	46.56	37.50	Anthropic	Proprietary
90	GPT-4.1-nano-2025-04-14 (Prompt)OpenAI	24.88	7.42	1.02	72.44	50.33	68.75	OpenAI	Proprietary
91	Falcon3-7B-Instruct (FC)TII UAE	24.03	73.61	93.11	82.69	68.32	100.00	TII UAE	falcon-llm-license
92	Qwen3-0.6B (FC)Qwen	23.93	0.46	0.68	71.79	56.62	75.00	Qwen	apache-2.0
93	Granite-20b-FunctionCalling (FC)IBM	23.23	5.23	3.20	82.35	58.70	87.50	IBM	Apache-2.0
94	Qwen3-0.6B (Prompt)Qwen	22.38	3.65	3.10	70.00	49.37	75.00	Qwen	apache-2.0
95	Amazon-Nova-Micro-v1:0 (FC)Amazon	22.29	1.81	1.12	74.10	66.32	81.25	Amazon	Proprietary
96	RZN-T (Prompt)Phronetic AI	22.25	12.31	12.32	67.94	49.74	81.25	Phronetic AI	apache-2.0
97	MiniCPM3-4B (Prompt)openbmb	22.08	29.83	31.18	70.54	43.15	56.25	openbmb	Apache-2.0
98	Llama-3.2-3B-Instruct (FC)Meta	21.95	6.20	6.10	82.67	58.33	87.50	Meta	Meta Llama 3 Community
99	Bielik-11B-v2.3-Instruct (Prompt)SpeakLeash & ACK Cyfronet AGH	21.90	22.44	23.75	81.50	67.80	93.75	SpeakLeash & ACK Cyfronet AGH	Apache 2.0
100	Hammer2.1-0.5b (FC)MadeAgents	21.22	2.82	2.79	65.98	54.63	68.75	MadeAgents	cc-by-nc-4.0
101	Gemma-3-4b-it (Prompt)Google	19.62	4.14	4.69	61.12	60.84	100.00	Google	gemma-terms-of-use
102	Open-Mistral-Nemo-2407 (Prompt)Mistral AI	19.31	13.80	0.84	88.46	73.95	93.75	Mistral AI	Proprietary
103	Granite-4.0-350m (FC)IBM	18.98	1.44	1.74	67.92	46.11	81.25	IBM	Apache-2.0
104	Falcon3-3B-Instruct (FC)TII UAE	16.25	36.70	38.52	54.62	54.48	81.25	TII UAE	falcon-llm-license
105	Ministral-8B-Instruct-2410 (FC)Mistral AI	11.10	70.01	82.07	—	—	—	Mistral AI	Mistral AI Research License
106	Falcon3-1B-Instruct (FC)TII UAE	11.08	1.72	5.23	9.02	2.89	—	TII UAE	falcon-llm-license
107	Llama-3.2-1B-Instruct (FC)Meta	10.82	1.64	3.21	38.38	11.77	43.75	Meta	Meta Llama 3 Community
108	Llama-3.1-Nemotron-Ultra-253B-v1 (FC)NVIDIA	10.00	0.72	1.42	—	—	—	NVIDIA	nvidia-open-model-license

数据仅供参考，以官方来源为准。模型名称旁的链接可跳转到 DataLearner 模型详情页。