Qwen2.5 32B Instruct
Alibaba Cloud / Qwen TeamQwenOpen WeightApache 2.0 · Commercial OK
描述
Qwen2.5-32B-Instruct is an instruction-tuned 32 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
32.5B
上下文长度
—
支持模态
—
能力雷达图
70
general
90
coding
80
reasoning
43
science估算
0
agents
0
multimodal
Science 在缺少专门科学评测时使用推理能力代理估算。
排行榜排名
| 领域 | #排名 | 分数 | 来源 |
|---|---|---|---|
| 推理能力 | 48 | 69.0 | LS |
基准测试分数 (LLM Stats)
Biology
GPQA
49.5%自报
Chemistry
MMLU-STEM
80.9%自报
Code
HumanEval
88.4%自报
Finance
MMLU
83.3%自报
MMLU-Pro
69.0%自报
TruthfulQA
57.8%自报
TheoremQA
44.1%自报
General
MBPP
0.84 / 100自报
MMLU-Redux
83.9%自报
MultiPL-E
75.4%自报
ARC-C
70.4%自报
MBPP+
67.2%自报
Language
BBH
84.5%自报
Winogrande
82.0%自报
Math
GSM8k
95.9%自报
MATH
83.1%自报
Reasoning
HellaSwag
85.2%自报
HumanEval+
52.4%自报
AA 评测指数
暂无 AA 评测数据
LLM Stats 分类评分
Code90
Language80
Math80
General70
Healthcare70
Legal70
Reasoning70
Finance60
Biology50
Chemistry50
Physics50
定价
暂无定价数据
速度
暂无速度数据
可用提供商
(LS 内部计价单位)暂无提供商数据