Text-to-Image
ZeroModels
Keras
PyTorch
JAX
TensorFlow
English
stable-diffusion
stable-diffusion-diffusers
stable-diffusion-xl
sdxl
diffusion
latent-diffusion
Instructions to use zeromodels/stable-diffusion-xl-refiner-1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ZeroModels
How to use zeromodels/stable-diffusion-xl-refiner-1.0 with ZeroModels:
# pip install -U zeromodels # ZeroModels is pure Keras 3, so pick a backend: "jax", "torch" or "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" from zeromodels import AutoZModel # AutoZModel reads the repo's model_type and loads the matching class. # For a task head use the matching loader, e.g. AutoZMImageClassify / AutoZMDetect / # AutoZMSemanticSegment / AutoZMTextGenerate (see zeromodels.auto). model = AutoZModel.from_weights("zeromodels/stable-diffusion-xl-refiner-1.0") - Keras
How to use zeromodels/stable-diffusion-xl-refiner-1.0 with Keras:
# !pip install -U keras tensorflow huggingface_hub # Keras needs TensorFlow installed to read "hf://" paths, so the tensorflow backend is selected here; # "jax" and "torch" also work for computation once TensorFlow is installed. import os os.environ["KERAS_BACKEND"] = "tensorflow" import keras model = keras.saving.load_model("hf://zeromodels/stable-diffusion-xl-refiner-1.0") - Notebooks
- Google Colab
- Kaggle
Add the attribution notice and the modification statement to the model card
Browse files
README.md
CHANGED
|
@@ -99,7 +99,7 @@ Load any Stable Diffusion XL checkpoint the same way with `from_weights("zeromod
|
|
| 99 |
| --- | --- | --- |
|
| 100 |
| `stable-diffusion-xl-base-1.0` | [zeromodels/stable-diffusion-xl-base-1.0](https://huggingface.co/zeromodels/stable-diffusion-xl-base-1.0) | 1024px, epsilon, Euler (leading spacing): the SDXL 1.0 base model, multi-aspect training with size / crop micro-conditioning |
|
| 101 |
| `stable-diffusion-xl-refiner-1.0` | [zeromodels/stable-diffusion-xl-refiner-1.0](https://huggingface.co/zeromodels/stable-diffusion-xl-refiner-1.0) | image-to-image refiner of SDXL 1.0: refines the base's latents (denoising_start 0.8) or an image (strength 0.3); one text tower, aesthetic-score conditioning |
|
| 102 |
-
| `sdxl-turbo` | [zeromodels/sdxl-turbo](https://huggingface.co/zeromodels/sdxl-turbo) | 512px, epsilon, ancestral Euler (trailing spacing), 1 to 4 steps, no guidance: SDXL 1.0 distilled with Adversarial Diffusion Distillation (
|
| 103 |
|
| 104 |
## Tips
|
| 105 |
|
|
@@ -119,9 +119,17 @@ Load any Stable Diffusion XL checkpoint the same way with `from_weights("zeromod
|
|
| 119 |
|
| 120 |
## License
|
| 121 |
|
| 122 |
-
The weights are redistributed under the
|
| 123 |
-
|
| 124 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 125 |
|
| 126 |
## Special Thanks
|
| 127 |
|
|
|
|
| 99 |
| --- | --- | --- |
|
| 100 |
| `stable-diffusion-xl-base-1.0` | [zeromodels/stable-diffusion-xl-base-1.0](https://huggingface.co/zeromodels/stable-diffusion-xl-base-1.0) | 1024px, epsilon, Euler (leading spacing): the SDXL 1.0 base model, multi-aspect training with size / crop micro-conditioning |
|
| 101 |
| `stable-diffusion-xl-refiner-1.0` | [zeromodels/stable-diffusion-xl-refiner-1.0](https://huggingface.co/zeromodels/stable-diffusion-xl-refiner-1.0) | image-to-image refiner of SDXL 1.0: refines the base's latents (denoising_start 0.8) or an image (strength 0.3); one text tower, aesthetic-score conditioning |
|
| 102 |
+
| `sdxl-turbo` | [zeromodels/sdxl-turbo](https://huggingface.co/zeromodels/sdxl-turbo) | 512px, epsilon, ancestral Euler (trailing spacing), 1 to 4 steps, no guidance: SDXL 1.0 distilled with Adversarial Diffusion Distillation (Stability AI Community License) |
|
| 103 |
|
| 104 |
## Tips
|
| 105 |
|
|
|
|
| 119 |
|
| 120 |
## License
|
| 121 |
|
| 122 |
+
The weights are redistributed under the [CreativeML Open RAIL++-M License](https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0/blob/main/LICENSE.md) of the upstream
|
| 123 |
+
checkpoint, including its use-based restrictions. By using them you agree to those terms.
|
| 124 |
+
|
| 125 |
+
## Notice
|
| 126 |
+
|
| 127 |
+
Modifications by zeromodels (https://github.com/IMvision12/ZeroModels): the checkpoint
|
| 128 |
+
released at https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0 was converted
|
| 129 |
+
to the Keras 3 weights layout of zeromodels (`model.weights.json, model_00000.weights.h5,
|
| 130 |
+
model_00001.weights.h5, zm_config.json, tokenizer.json`), stored in float16, the upstream
|
| 131 |
+
fp16 files, with the VAE in float32. The model architecture and the parameter values are
|
| 132 |
+
unchanged; the weight names and the file format differ from the release.
|
| 133 |
|
| 134 |
## Special Thanks
|
| 135 |
|