Spaces:
Runtime error
Runtime error
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,13 +1,100 @@
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: gradio
|
| 7 |
-
sdk_version:
|
| 8 |
-
python_version: '3.12'
|
| 9 |
app_file: app.py
|
| 10 |
pinned: false
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
title: Z-Image Turbo Image to Image
|
| 3 |
+
emoji: 🖼️
|
| 4 |
+
colorFrom: blue
|
| 5 |
+
colorTo: indigo
|
| 6 |
sdk: gradio
|
| 7 |
+
sdk_version: 5.49.1
|
|
|
|
| 8 |
app_file: app.py
|
| 9 |
pinned: false
|
| 10 |
+
hardware: zerogpu
|
| 11 |
---
|
| 12 |
|
| 13 |
+
# Z-Image Turbo — Image to Image
|
| 14 |
+
|
| 15 |
+
A simple image-to-image generation Space powered by
|
| 16 |
+
**Tongyi-MAI/Z-Image-Turbo** and Hugging Face ZeroGPU.
|
| 17 |
+
|
| 18 |
+
## Features
|
| 19 |
+
|
| 20 |
+
- Upload an existing image
|
| 21 |
+
- Describe the desired transformation
|
| 22 |
+
- Control transformation strength
|
| 23 |
+
- Use a fixed seed for reproducible results
|
| 24 |
+
- Random seed mode
|
| 25 |
+
- 8-step Z-Image-Turbo generation
|
| 26 |
+
- Mobile-friendly Gradio interface
|
| 27 |
+
- ZeroGPU acceleration
|
| 28 |
+
|
| 29 |
+
## Model
|
| 30 |
+
|
| 31 |
+
This Space uses:
|
| 32 |
+
|
| 33 |
+
`Tongyi-MAI/Z-Image-Turbo`
|
| 34 |
+
|
| 35 |
+
Z-Image-Turbo is a 6B-parameter image generation model designed for efficient generation. Its image-to-image pipeline is provided through Diffusers.
|
| 36 |
+
|
| 37 |
+
## How it works
|
| 38 |
+
|
| 39 |
+
1. Upload an image.
|
| 40 |
+
2. Enter a prompt describing the transformation.
|
| 41 |
+
3. Adjust transformation strength.
|
| 42 |
+
4. Click **Generate Image**.
|
| 43 |
+
5. ZeroGPU allocates GPU compute for the generation.
|
| 44 |
+
6. The generated image is returned to the interface.
|
| 45 |
+
|
| 46 |
+
## Transformation Strength
|
| 47 |
+
|
| 48 |
+
### 0.1–0.3
|
| 49 |
+
|
| 50 |
+
Preserves most of the original image.
|
| 51 |
+
|
| 52 |
+
Good for:
|
| 53 |
+
|
| 54 |
+
- Lighting changes
|
| 55 |
+
- Color changes
|
| 56 |
+
- Minor style changes
|
| 57 |
+
|
| 58 |
+
### 0.4–0.7
|
| 59 |
+
|
| 60 |
+
Balanced transformation.
|
| 61 |
+
|
| 62 |
+
Good for:
|
| 63 |
+
|
| 64 |
+
- Art style changes
|
| 65 |
+
- Environment changes
|
| 66 |
+
- Character/environment redesign
|
| 67 |
+
|
| 68 |
+
### 0.8–1.0
|
| 69 |
+
|
| 70 |
+
Strong transformation.
|
| 71 |
+
|
| 72 |
+
The generated image can differ substantially from the original.
|
| 73 |
+
|
| 74 |
+
## ZeroGPU
|
| 75 |
+
|
| 76 |
+
This Space is designed for Hugging Face ZeroGPU.
|
| 77 |
+
|
| 78 |
+
The GPU is requested only while the generation function is running.
|
| 79 |
+
|
| 80 |
+
Free ZeroGPU availability is subject to Hugging Face account eligibility, quotas, and queue availability.
|
| 81 |
+
|
| 82 |
+
## License
|
| 83 |
+
|
| 84 |
+
The application code in this Space is provided as-is.
|
| 85 |
+
|
| 86 |
+
The model is provided by:
|
| 87 |
+
|
| 88 |
+
https://huggingface.co/Tongyi-MAI/Z-Image-Turbo
|
| 89 |
+
|
| 90 |
+
Please check the model's license and terms before deploying or redistributing it.
|
| 91 |
+
|
| 92 |
+
## Credits
|
| 93 |
+
|
| 94 |
+
Built with:
|
| 95 |
+
|
| 96 |
+
- Hugging Face Spaces
|
| 97 |
+
- Gradio
|
| 98 |
+
- Diffusers
|
| 99 |
+
- PyTorch
|
| 100 |
+
- Z-Image-Turbo
|