echarlaix HF staff commited on
Commit
1b47b10
1 Parent(s): 43d0fa6

add optimum examples

Browse files
Files changed (1) hide show
  1. README.md +52 -1
README.md CHANGED
@@ -88,6 +88,57 @@ instead of `.to("cuda")`:
88
  ```
89
 
90
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
91
  ## Uses
92
 
93
  ### Direct Use
@@ -117,4 +168,4 @@ The model was not trained to be factual or true representations of people or eve
117
  - The autoencoding part of the model is lossy.
118
 
119
  ### Bias
120
- While the capabilities of image generation models are impressive, they can also reinforce or exacerbate social biases.
 
88
  ```
89
 
90
 
91
+ ### Optimum
92
+ [Optimum](https://github.com/huggingface/optimum) provides a Stable Diffusion pipeline compatible with both [OpenVINO](https://docs.openvino.ai/latest/index.html) and [ONNX Runtime](https://onnxruntime.ai/).
93
+
94
+ #### OpenVINO
95
+
96
+ To install Optimum with the dependencies required for OpenVINO :
97
+
98
+ ```bash
99
+ pip install optimum[openvino]
100
+ ```
101
+
102
+ To load an OpenVINO model and run inference with OpenVINO Runtime, you need to replace `StableDiffusionXLPipeline` with Optimum `OVStableDiffusionXLPipeline`. In case you want to load a PyTorch model and convert it to the OpenVINO format on-the-fly, you can set `export=True`.
103
+
104
+ ```diff
105
+ - from diffusers import StableDiffusionPipeline
106
+ + from optimum.intel import OVStableDiffusionPipeline
107
+
108
+ model_id = "stabilityai/stable-diffusion-xl-base-1.0"
109
+ - pipeline = StableDiffusionPipeline.from_pretrained(model_id)
110
+ + pipeline = OVStableDiffusionPipeline.from_pretrained(model_id)
111
+ prompt = "A majestic lion jumping from a big stone at night"
112
+ image = pipeline(prompt).images[0]
113
+ ```
114
+
115
+ You can find more examples (such as static reshaping and model compilation) in optimum [documentation](https://huggingface.co/docs/optimum/main/en/intel/inference#stable-diffusion-xl).
116
+
117
+
118
+ #### ONNX
119
+
120
+ To install Optimum with the dependencies required for ONNX Runtime inference :
121
+
122
+ ```bash
123
+ pip install optimum[onnxruntime]
124
+ ```
125
+
126
+ To load an ONNX model and run inference with ONNX Runtime, you need to replace `StableDiffusionXLPipeline` with Optimum `ORTStableDiffusionXLPipeline`. In case you want to load a PyTorch model and convert it to the ONNX format on-the-fly, you can set `export=True`.
127
+
128
+ ```diff
129
+ - from diffusers import StableDiffusionPipeline
130
+ + from optimum.onnxruntime import ORTStableDiffusionPipeline
131
+
132
+ model_id = "stabilityai/stable-diffusion-xl-base-1.0"
133
+ - pipeline = StableDiffusionPipeline.from_pretrained(model_id)
134
+ + pipeline = ORTStableDiffusionPipeline.from_pretrained(model_id)
135
+ prompt = "A majestic lion jumping from a big stone at night"
136
+ image = pipeline(prompt).images[0]
137
+ ```
138
+
139
+ You can find more examples in optimum [documentation](https://huggingface.co/docs/optimum/main/en/onnxruntime/usage_guides/models#stable-diffusion-xl).
140
+
141
+
142
  ## Uses
143
 
144
  ### Direct Use
 
168
  - The autoencoding part of the model is lossy.
169
 
170
  ### Bias
171
+ While the capabilities of image generation models are impressive, they can also reinforce or exacerbate social biases.