ChatGPT智能语言大模型是基于深度学习神经网络ToKens参数

原创作者: 太极混元发布时间： 2024-04-22 3.09K 人阅读

AI智能语言模型是基于深度学习技术，采用深度神经网络进行建模和训练。模型表准备更新中…

全局调用限制

360智脑 API 开放平台会对你可以向 API 发送的请求实施速率限制。这些限制适用于每分钟请求数（RPM）、每分钟令牌数（TPM），或者在图像模型的情况下每分钟调用次数。

模型名称	TPM(每分钟TOKEN)	RPM(每分钟请求数)	价格（模型输入）	价格（模型输出）
360gpt-turbo-responsibility-8k	120,000	300	¥0.012 / 1K tokens	¥0.012 / 1K tokens
360gpt-pro	120,000	300	¥0.012 / 1K tokens	¥0.012 / 1K tokens
360gpt-turbo	120,000	300	¥0.0012 / 1K tokens	¥0.0012 / 1K tokens
embedding-bert-512-v1	25,000	3,000	¥0.001 / 1K tokens	–
embedding_s1_v1.2	5,000,000	3,000	¥0.001 / 1K tokens	–
360CV_S0_V5	–	100	–	¥0.1 / images

同一组织内多个账号的总速率不能超出该组织的调用限制。您还可以设定组织账户配额，以避免账户资金被滥用。

OpenCompass Large Language Model Leaderboard

系列号	Model	Release	Type	Parameters	Average	Language	知识	Reasoning	Math	Code	Agent

	模型名称随机
1	GPT-4-Turbo-1106 OpenAI	2023/11/6updated: 2024/1/29	Chat	N/A	64.4	58.6	79.2	43.9	53.5	67.9	83.3
2	Claude3-Opus Anthropic	2024/3/4updated: 2024/4/2	Chat	N/A	65.3	52.6	83.5	42.2	62.5	69.6	81.6
3	GLM-4 ZhipuAI	2024/1/16updated: 2024/4/2	Chat	N/A	57.5	59.2	75	34.3	46.7	53.3	76.5
4	Qwen-Max-0107 Alibaba	2023/12/1updated: 2024/1/29	Chat	N/A	55.7	56	79.9	30.3	44.1	57.7	66.4
5	Qwen-Max-0403 Alibaba	2024/3/26updated: 2024/4/2	Chat	N/A	57.3	56.2	76.2	43.8	36.3	54.6	76.6
6	Qwen1.5-72B-Chat Alibaba	2024/2/4updated: 2024/2/20	Chat	72B	54.8	57.5	76.8	32.7	50.9	52.6	58.3
7	Erniebot-4.0 Baidu Inc.	2023/10/18updated: 2024/1/29	Chat	N/A	55.6	56.6	72.2	36.8	50.4	57.2	60.6
8	UniGPT Unisound	2023/8/28updated: 2024/4/2	Chat	N/A	52.5	57.1	77	42.1	35.3	42.7	60.7
9	Mistral-Large Mistral AI	2024/2/26updated: 2024/4/2	Chat	N/A	57.5	52.8	82.9	30	48.7	49.2	81.5
10	Qwen-72B-Chat Alibaba	2023/11/30updated: 2024/1/29	Chat	72B	51.2	49.4	76.7	27.1	45.5	50.3	58.1
11	MiniMax-abab5.5 MiniMax	2023/10/19updated: 2024/1/29	Chat	N/A	53.4	56.8	76.4	45	31	47.2	64
12	Qwen1.5-14B-Chat Alibaba	2024/2/4updated: 2024/2/20	Chat	14B	48.7	55.4	61.4	25.4	44.2	41.7	64.2
13	InternLM2-Chat-20B Shanghai AI Lab	2024/1/11updated: 2024/1/29	Chat	20B	52.3	58.8	69.2	30	46.3	54.9	54.5
14	Yi-34B-Chat 01.AI	2023/11/22updated: 2024/1/29	Chat	34B	47.9	48.6	76.2	37.4	37.1	34.5	53.8
15	GPT-3.5-Turbo OpenAI	2023/6/13updated: 2024/1/29	Chat	N/A	52	46.2	74	22.6	35.6	59.3	74.2
16	OrionStar-Yi-34B-Chat OrionStarAI	2023/11/16updated: 2023/11/22	Chat	34B	45.8	54.4	75	39.1	32.1	33.2	40.7
17	Baichuan2-Turbo Baichuan Intelligent Technology	2023/12/19updated: 2024/1/29	Chat	N/A	47.4	37.3	73.7	28.9	40.1	41.8	62.5
18	DBRX-Instruct DataBricks	2024/3/26updated: 2024/4/2	Chat	132B	50.6	46.7	81.9	25.2	38.9	51.9	58.9
19	InternLM2-Chat-7B Shanghai AI Lab	2024/1/11updated: 2024/1/29	Chat	7B	48.8	55.2	65.8	35	35.3	49.1	52.6
20	Qwen-14B-Chat Alibaba	2023/9/25updated: 2024/1/29	Chat	14B	44.6	47.1	68.9	28.2	37.8	31.7	54.1
21	DeepSeek-67B-Chat DeepSeek	2023/11/29updated: 2024/1/29	Chat	67B	45.5	30.1	72.2	25.6	36.4	53.5	54.8
22	Qwen1.5-7B-Chat Alibaba	2024/2/4updated: 2024/2/20	Chat	7B	39.8	43.9	53.5	21.3	30.7	36.7	52.8
23	Nanbeige2-8B-Chat Nanbeige	2024/3/24updated: 2024/4/2	Chat	8B	41.7	44.3	61.9	23.9	30.3	32.6	57.4
24	Nanbeige-16B-Chat Nanbeige	2023/11/8updated: 2024/1/29	Chat	16B	40.9	52.9	61	32.8	17.3	31.4	49.9
25	Mixtral-8x7B-Instruct-v0.1 Mistral AI	2023/12/11updated: 2024/1/29	Chat	47B	43.5	44.9	76.5	20.2	42.3	26.1	50.7
26	Qwen-7B-Chat Alibaba	2023/8/3updated: 2024/1/29	Chat	7B	38.2	39.8	61.7	22.6	25.5	30	49.7
27	ChatGLM3-6B-32K ZhipuAI	2023/10/27updated: 2024/1/29	Chat	6B	37	43.9	54.4	19.5	23.8	39.2	41
28	Yi-6B-Chat 01.AI	2023/11/22updated: 2024/1/29	Chat	6B	32.3	34.6	62.9	18.9	17.7	17.5	41.9
29	Baichuan2-13B-Chat Baichuan Intelligent Technology	2023/9/6updated: 2024/1/29	Chat	13B	35.9	40.4	62.5	29.4	19.9	25.2	37.8
30	WizardLM-70B-V1.0 Microsoft	2023/8/9updated: 2024/1/29	Chat	70B	31.5	19.7	67.8	9.7	29.4	24.8	37.8
31	LLaMA-2-70B-Chat Meta	2023/7/19updated: 2024/1/29	Chat	70B	41	49.1	75.2	28.1	26.3	24.6	42.5
32	DeepSeek-7B-Chat DeepSeek	2023/11/29updated: 2024/1/29	Chat	7B	29.4	22	45.2	9.2	17.4	37.8	44.8
33	Baichuan2-7B-Chat Baichuan Intelligent Technology	2023/9/6updated: 2024/1/29	Chat	7B	31	33.7	55.8	20.5	12.7	22.6	41
34	Mistral-7B-Instruct-v0.2 Mistral AI	2023/12/11updated: 2024/1/29	Chat	7B	34.5	39.4	67.9	20.4	19.6	20.9	38.5
35	Vicuna-13B-v1.5-16k LMSYS	2023/7/31updated: 2024/1/29	Chat	13B	32.5	34.3	70.7	28.8	12.7	6.4	42.3
36	WizardLM-13B-V1.2 Microsoft	2023/7/25updated: 2024/1/29	Chat	13B	32.2	32.1	59	13.7	19.2	27.4	42
37	Zephyr-7B-β HuggingFace	2023/10/26updated: 2024/1/29	Chat	7B	31.1	28.4	66.8	14.2	16.1	22.6	38.3
38	LLaMA-2-13B-Chat Meta	2023/7/19updated: 2024/1/29	Chat	13B	32	37.4	69.3	23.3	15	13.8	33.2
39	Vicuna-7B-v1.5-16k LMSYS	2023/8/7updated: 2024/1/29	Chat	7B	24.8	13.8	64.5	13.9	8	8.3	40.1
40	LLaMA-2-7B-Chat Meta	2023/7/19updated: 2024/1/29	Chat	7B	27.8	27.7	61.2	22.2	8.9	17.9	28.6

我们在OpenCompass的主流评测数据集上验证了我们的模型性能，包括C-Eval、AGIEval、MMLU、CMMLU、HellaSwag、MATH、GSM8K、HumanEval、MBPP、BBH、LAMBADA，考察的能力包括自然语言理解、知识、数学计算和推理、代码生成、逻辑推理等。

Model	AVG	CEval	AGIEval	MMLU	CMMLU	HellaSwag	MATH	GSM8K	人类	MBPP	BBH	LAMBADA
Baichuan2-7B	41.49	56.3	34.6	54.7	57	67	5.4	24.6	17.7	24	41.8	73.3
Baichuan-7B	31.94	44.7	24.6	41.5	44.6	68.4	2.5	9.6	9.1	6.4	32.8	67.1
ChatGLM3-6B	58.67	67	47.4	62.8	66.5	76.5	19.2	61	44.5	57.2	66.2	77.1
DeepSeek-7B	39.8	45	24	49.3	46.8	73.4	4.2	18.3	25	36.4	42.8	72.6
InternLM2-7B	58.01	65.7	50.2	65.5	66.2	79.6	19.9	70.6	41.5	42.4	64.4	72.1
InternLM-7B	39.33	53.4	36.9	51	51.8	70.6	6.3	31.2	13.4	14	37	67
LLaMA-2-7B	33.27	32.5	21.8	46.8	31.8	74	3.3	16.7	12.8	14.8	38.2	73.3
LLaMA-7B	30.35	27.3	20.6	35.6	26.8	74.3	2.9	10	12.8	16.8	33.5	73.3
Mistral-7B-v0.1	47.67	47.4	32.8	64.1	44.7	78.9	11.3	47.5	27.4	38.6	56.7	75
MPT-7B	30.06	23.5	21.3	27.5	25.9	75	2.9	9.1	17.1	22.8	35.6	70
Qwen1.5-7B	55.12	73.57	50.8	62.15	71.84	72.62	20.36	54.36	53.05	36.8	40.01	70.74
Qwen-7B	49.53	63.4	45.3	59.7	62.5	75	13.3	54.1	27.4	31.4	45.2	67.5
XVERSE-7B	34.27	61.1	39	58.4	60.8	73.7	2.2	11.7	4.9	10.2	31	24
Yi-6B	47.8	73	44.3	64	73.5	73.1	6.3	39.9	15.2	23.6	44.9	68
360Zhinao-7B	56.15	74.11	49.49	67.44	72.38	83.05	16.38	53.83	35.98	42.4	43.95	78.59

以上结果，在官方Opencompass上可查询或可复现。

全局调用限制

提供中小企业建站高端正版精品系统

工作时间