Update README.md
Browse files
README.md
CHANGED
@@ -17,7 +17,7 @@ Haoyu Lu*, Wen Liu*, Bo Zhang**, Bingxuan Wang, Kai Dong, Bo Liu, Jingxiang Sun,
|
|
17 |
|
18 |
### 2. Model Summary
|
19 |
|
20 |
-
DeepSeek-VL-1.3b is a tiny vision-language model. It uses the [SigLIP-L](https://huggingface.co/timm/ViT-L-16-SigLIP-384) as the vision encoder supporting 384 x 384 image input
|
21 |
and is constructed based on the DeepSeek-LLM-1.3b-base which is trained on an approximate corpus of 500B text tokens. The whole DeepSeek-1.3b-base model is finally trained around 400B vision-language tokens.
|
22 |
|
23 |
## 3. Quick Start
|
|
|
17 |
|
18 |
### 2. Model Summary
|
19 |
|
20 |
+
DeepSeek-VL-1.3b-base is a tiny vision-language model. It uses the [SigLIP-L](https://huggingface.co/timm/ViT-L-16-SigLIP-384) as the vision encoder supporting 384 x 384 image input
|
21 |
and is constructed based on the DeepSeek-LLM-1.3b-base which is trained on an approximate corpus of 500B text tokens. The whole DeepSeek-1.3b-base model is finally trained around 400B vision-language tokens.
|
22 |
|
23 |
## 3. Quick Start
|