deepseekmirror/DeepSeek-VL2
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepSeek-VL2.git synced 2025-04-19 18:19:03 -04:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge 3381e67616 into ef9f91e2b6

Browse Source
This commit is contained in:
ved pawar 2025-02-26 13:18:26 +08:00 committed by GitHub
commit 1c21d366d5
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 10 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
README.md
View File

@ -56,6 +56,16 @@
<a href="https://huggingface.co/spaces/deepseek-ai/deepseek-vl2-small"><b>👁️ Demo</b></a> <a href="https://huggingface.co/spaces/deepseek-ai/deepseek-vl2-small"><b>👁️ Demo</b></a>
</p> </p>
## Table of Contents
1. [Introduction](#1-introduction)
2. [Release](#2-release)
3. [Model Download](#3-model-download)
4. [Quick Start](#4-quick-start)
5. [License](#5-license)
6. [Citation](#6-citation)
## 1. Introduction ## 1. Introduction
Introducing DeepSeek-VL2, 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. Our model series is composed of three variants: DeepSeek-VL2-Tiny, DeepSeek-VL2-Small and DeepSeek-VL2, with 1.0B, 2.8B and 4.5B activated parameters respectively. Introducing DeepSeek-VL2, 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. Our model series is composed of three variants: DeepSeek-VL2-Tiny, DeepSeek-VL2-Small and DeepSeek-VL2, with 1.0B, 2.8B and 4.5B activated parameters respectively.