DeepSeek VL2 Tiny
DeepSeekDeepSeekOpen Weightdeepseek
描述
An advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL. DeepSeek-VL2 demonstrates superior capabilities across various tasks, including but not limited to visual question answering, optical character recognition, document/table/chart understanding, and visual grounding.
发布日期
2024-12-13
参数规模
3.0B
上下文长度
164K
支持模态
text
能力雷达图
50
general
0
coding
50
reasoning
34
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 |
|---|---|---|---|
| 多模态榜 | 63 | 69.0 | LS |
基准测试分数 (LLM Stats)
General
MMT-Bench
53.2%自报
MMStar
45.9%自报
MMMU
40.7%自报
Image To Text
DocVQA
88.9%自报
OCRBench
80.9%自报
TextVQA
80.7%自报
Math
MathVista
53.6%自报
Multimodal
ChartQA
81.0%自报
AI2D
71.6%自报
MMBench
69.2%自报
MMBench-V1.1
68.3%自报
InfoVQA
66.1%自报
MME
19.1%自报
Spatial Reasoning
RealWorldQA
64.2%自报
AA 评测指数
暂无 AA 评测数据
LLM Stats 分类评分
Image To Text80
Spatial Reasoning60
Vision60
Multimodal60
Reasoning60
General50
Math50
Healthcare40
定价
输入价格$0.32 / 1M tokens
输出价格$0.89 / 1M tokens
混合价格(3:1)$0.4625 / 1M tokens
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