Qwen2.5 Coder Instruct 32B
AlibabaQwenOpen WeightApache 2.0 · Commercial OK
描述
Qwen2.5-Coder is a specialized coding model trained on 5.5 trillion tokens of code data, supporting 92 programming languages with a 128K context window. It excels in code generation, completion, repair, and multi-programming tasks while maintaining strong performance in mathematics and general capabilities.
发布日期
2024-11-11
参数规模
32.0B
上下文长度
33K
支持模态
text
能力雷达图
27
general
29
coding
37
reasoning
29
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
基准测试分数 (LLM Stats)
Code
HumanEval
92.7%自报
LiveCodeBench
31.4%自报
Finance
MMLU
75.1%自报
TruthfulQA
54.2%自报
MMLU-Pro
50.4%自报
TheoremQA
43.1%自报
General
MBPP
0.90 / 100自报
MMLU-Redux
77.5%自报
ARC-C
70.5%自报
BigCodeBench-Full
49.6%自报
BigCodeBench-Hard
27.0%自报
Language
Winogrande
80.8%自报
Math
GSM8k
91.1%自报
MATH
57.2%自报
Reasoning
HellaSwag
83.0%自报
AA 评测指数
Intelligence Index12.9
Math 5000.8
Mmlu Pro0.6
Gpqa0.4
Livecodebench0.3
Scicode0.3
Aime0.1
Hle0.0
LLM Stats 分类评分
Language70
Math70
Reasoning70
Code60
Finance60
General60
Healthcare60
Legal60
Physics40
定价
输入价格免费
输出价格免费
混合价格(3:1)免费
速度
Tokens/秒0.0 tokens/s
首Token延迟0.00s
首回答延迟0.00s
可用提供商
(LS 内部计价单位)暂无提供商数据