跳转到主要内容
AITier

Qwen2.5 Instruct 72B

AlibabaQwen开源权重Qwen · 商用许可

描述

Qwen2.5-72B-Instruct is an instruction-tuned 72 billion parameter language model, part of the Qwen2.5 series. It is designed to follow instructions, generate long texts (over 8K tokens), understand structured data (e.g., tables), and generate structured outputs, especially JSON. The model supports multilingual capabilities across over 29 languages.

发布日期
2024-09-19
参数规模
72.7B
上下文长度
131K
支持模态
text

能力雷达图

27
general
27
coding
29
reasoning
32
science
28
agents
0
multimodal

排行榜排名

领域#排名分数来源
代码能力榜428
20.0
AA
通用能力榜373
35.0
AA
科学能力421
32.0
AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Biology

GPQANYU + Cohere + Anthropic (2023)49.0%自报

Code

HumanEvalOpenAI (2021)86.6%自报
LiveCodeBench55.5%自报

Communication

MT-Bench0.94 / 100自报

Creativity

AlignBench81.6%自报
Arena Hard81.2%自报

Finance

MMLU-Pro71.1%自报

General

MBPP0.88 / 100自报
MMLU-Redux86.8%自报
IFEvalGoogle Research (2023)84.1%自报
MultiPL-E75.1%自报
LiveBench52.3%自报

Math

GSM8k95.8%自报
MATH83.1%自报

AA 评测指数

(Artificial Analysis)
Math Index(Artificial Analysis)
14.0
Intelligence Index(Artificial Analysis)
9.4
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))
0.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.7
Gpqa(NYU + Cohere + Anthropic (2023))
0.5
Ifbench(Google Research (2023))
0.4
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.3
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.3
Scicode(UIUC + Argonne National Lab (2024))
0.3
Lcr(Artificial Analysis)
0.2
Aime(MAA (Mathematical Association of America))
0.2
Aime 25(MAA (Mathematical Association of America))
0.1
Terminalbench Hard(Stanford × Laude Institute (2026))
0.0
Hle(Center for AI Safety + Scale AI (2025))
0.0

LLM Stats 分类评分

(LLM Stats (zeroeval))
Roleplay
90
Communication
90
Creativity
90
Math
80
Reasoning
80
Structured Output
80
Instruction Following
80
Language
80
General
80
Writing
80
Legal
70
Finance
70
Healthcare
70
Code
70
Physics
50
Biology
50
Chemistry
50

定价

输入价格$0.475 / 1M tokens
输出价格$0.495 / 1M tokens
混合价格(3:1)$0.48 / 1M tokens

速度

Tokens/秒0.0
首Token延迟0.00s
首回答延迟0.00s

供应商价格排行

供应商价格排行

2 个供应商

最便宜: Alibaba最贵: Alibaba (China)
供应商输入输出
1Alibaba主要
$0.475
$0.495
2Alibaba (China)
$0.574
$1.721

比较该模型在不同 API 供应商之间的定价。

外部链接

LLM StatsArtificial Analysis