Spaces:
Running
on
Zero
Running
on
Zero
Merge branch 'main' of https://huggingface.co/spaces/deepseek-ai/Janus-Pro-7B into main
Browse files- README.md +3 -1
- requirements.txt +1 -0
README.md
CHANGED
|
@@ -11,4 +11,6 @@ license: mit
|
|
| 11 |
short_description: A unified multimodal understanding and generation model.
|
| 12 |
---
|
| 13 |
|
| 14 |
-
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
|
|
|
|
|
|
|
|
| 11 |
short_description: A unified multimodal understanding and generation model.
|
| 12 |
---
|
| 13 |
|
| 14 |
+
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
| 15 |
+
|
| 16 |
+
Please note that after generating the images with Janus-Pro, we used Real-ESRGAN to perform 2x image upsampling. This upsampling does not alter the content of the images but makes them slightly clearer for easier viewing.
|
requirements.txt
CHANGED
|
@@ -5,4 +5,5 @@ numpy
|
|
| 5 |
torch
|
| 6 |
safetensors
|
| 7 |
transformers
|
|
|
|
| 8 |
git+https://github.com/deepseek-ai/Janus
|
|
|
|
| 5 |
torch
|
| 6 |
safetensors
|
| 7 |
transformers
|
| 8 |
+
opencv-python
|
| 9 |
git+https://github.com/deepseek-ai/Janus
|