deepseek-community/deepseek-vl-1.3b-chat
Image-Text-to-Text • 2B • Updated • 3.3k • 3
Deepseek-VL
Deepseek Vision-Language Models - Original and Transformers Compatible Weights
-
-
deepseek-community/deepseek-vl-7b-chat
Image-Text-to-Text • 7B • Updated • 3.3k • 1 -
deepseek-community/deepseek-vl-1.3b-base
Image-Text-to-Text • 2B • Updated • 7
deepseek-community/deepseek-vl-7b-base
Image-Text-to-Text • 7B • Updated • 34Note ^ Transformers version, compatible with the image-text-to-text task.
-
deepseek-ai/deepseek-vl-1.3b-chat
Image-Text-to-Text • 2B • Updated • 12.6k • 71 -
deepseek-ai/deepseek-vl-7b-chat
Image-Text-to-Text • 7B • Updated • 2.16k • 270 -
deepseek-ai/deepseek-vl-1.3b-base
2B • Updated • 205 • 56
deepseek-ai/deepseek-vl-7b-base
7B • Updated • 153 • 65Note ^ Original checkpoint, compatible with the research codebase.
-
DeepSeek-VL: Towards Real-World Vision-Language Understanding
Paper • 2403.05525 • Published • 49