Instructions to use PANDATREE/BRIDGE with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use PANDATREE/BRIDGE with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("PANDATREE/BRIDGE", dtype=torch.bfloat16, device_map="cuda") prompt = "Turn this cat into a dog" input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cat.png") image = pipe(image=input_image, prompt=prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Add pipeline tag, library name, and paper link
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,18 +1,22 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
| 2 |
license: other
|
| 3 |
license_name: mixed-see-licensing
|
| 4 |
license_link: https://huggingface.co/PANDATREE/BRIDGE/blob/main/README.md#licensing
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
- black-forest-labs/FLUX.2-klein-base-9B
|
| 8 |
tags:
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
---
|
| 13 |
|
| 14 |
# BRIDGE Model Weights
|
| 15 |
|
|
|
|
|
|
|
| 16 |
BRIDGE uses a main generation path, an independently denoised subject path, and
|
| 17 |
learned discrete positional-encoding routing. This repository now contains both
|
| 18 |
the original Qwen release and two FLUX.2 Klein 9B variants. They are different
|
|
@@ -80,4 +84,4 @@ must not be applied to the new FLUX weights.
|
|
| 80 |
journal={arXiv preprint arXiv:2605.07846},
|
| 81 |
year={2025}
|
| 82 |
}
|
| 83 |
-
```
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model:
|
| 3 |
+
- Qwen/Qwen-Image-Edit-2511
|
| 4 |
+
- black-forest-labs/FLUX.2-klein-base-9B
|
| 5 |
license: other
|
| 6 |
license_name: mixed-see-licensing
|
| 7 |
license_link: https://huggingface.co/PANDATREE/BRIDGE/blob/main/README.md#licensing
|
| 8 |
+
pipeline_tag: image-to-image
|
| 9 |
+
library_name: diffusers
|
|
|
|
| 10 |
tags:
|
| 11 |
+
- image-editing
|
| 12 |
+
- subject-driven-generation
|
| 13 |
+
- positional-encoding
|
| 14 |
---
|
| 15 |
|
| 16 |
# BRIDGE Model Weights
|
| 17 |
|
| 18 |
+
The model is described in the paper [BRIDGE: Background Routing and Isolated Discrete Gating for Coarse-Mask Local Editing](https://huggingface.co/papers/2605.07846). Code is available at [the BRIDGE repository](https://github.com/XIONGPEILIN/BRIDGE).
|
| 19 |
+
|
| 20 |
BRIDGE uses a main generation path, an independently denoised subject path, and
|
| 21 |
learned discrete positional-encoding routing. This repository now contains both
|
| 22 |
the original Qwen release and two FLUX.2 Klein 9B variants. They are different
|
|
|
|
| 84 |
journal={arXiv preprint arXiv:2605.07846},
|
| 85 |
year={2025}
|
| 86 |
}
|
| 87 |
+
```
|