Instructions to use neuralvfx/LibreFlux-SAM-ControlNet with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use neuralvfx/LibreFlux-SAM-ControlNet with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("neuralvfx/LibreFlux-SAM-ControlNet", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Update README.md
Browse files
README.md
CHANGED
|
@@ -10,14 +10,12 @@ base_model: jimmycarter/LibreFLUX
|
|
| 10 |

|
| 11 |
|
| 12 |
# Fun Facts
|
| 13 |
-
-
|
| 14 |
-
- This ControlNet is trained exclusively on images generated by [Segment Anything (SAM)](https://aidemos.meta.com/segment-anything/)
|
| 15 |
- Uses SAM style images as input, outputs photorealistic images
|
| 16 |
- Trained at 1024x1024 resolution, inference works best at 1.5k and up
|
| 17 |
- Trained on 320K segmented images from [laion2b-squareish-1536px](https://huggingface.co/datasets/opendiffusionai/laion2b-squareish-1536px)
|
| 18 |
-
- Base
|
| 19 |
-
|
| 20 |
-
|
| 21 |
|
| 22 |
# Showcases
|
| 23 |
<table style="width:100%; table-layout:fixed;">
|
|
@@ -36,11 +34,13 @@ base_model: jimmycarter/LibreFLUX
|
|
| 36 |
</table>
|
| 37 |
|
| 38 |
|
| 39 |
-
#
|
| 40 |
-
-
|
| 41 |
- Trained in same non-distilled fashion
|
| 42 |
- Uses Attention Masking
|
| 43 |
-
- Uses CFG during Inference
|
|
|
|
|
|
|
| 44 |
|
| 45 |
# Compatibility
|
| 46 |
```py
|
|
|
|
| 10 |

|
| 11 |
|
| 12 |
# Fun Facts
|
| 13 |
+
- Trained exclusively on images generated by [Segment Anything (SAM)](https://aidemos.meta.com/segment-anything/)
|
|
|
|
| 14 |
- Uses SAM style images as input, outputs photorealistic images
|
| 15 |
- Trained at 1024x1024 resolution, inference works best at 1.5k and up
|
| 16 |
- Trained on 320K segmented images from [laion2b-squareish-1536px](https://huggingface.co/datasets/opendiffusionai/laion2b-squareish-1536px)
|
| 17 |
+
- Base model is [LibreFLUX](https://huggingface.co/jimmycarter/LibreFLUX)
|
| 18 |
+
|
|
|
|
| 19 |
|
| 20 |
# Showcases
|
| 21 |
<table style="width:100%; table-layout:fixed;">
|
|
|
|
| 34 |
</table>
|
| 35 |
|
| 36 |
|
| 37 |
+
# Extra Details
|
| 38 |
+
- I built this repo to train the model: [https://github.com/NeuralVFX/LibreFLUX-ControlNet](https://github.com/NeuralVFX/LibreFLUX-ControlNet)
|
| 39 |
- Trained in same non-distilled fashion
|
| 40 |
- Uses Attention Masking
|
| 41 |
+
- Uses CFG during Inference ( allows negative prompting )
|
| 42 |
+
- Inference code roughly adapted from: [https://github.com/bghira/SimpleTuner](https://github.com/bghira/SimpleTuner)
|
| 43 |
+
|
| 44 |
|
| 45 |
# Compatibility
|
| 46 |
```py
|