eyal.benaroche commited on
Commit
ebec059
1 Parent(s): d115e17

update Readme with comfy instructions

Browse files
Files changed (1) hide show
  1. README.md +12 -0
README.md CHANGED
@@ -98,6 +98,18 @@ image = pipe(
98
  <img style="width:400px;" src="images/corgi.jpg">
99
  </p>
100
 
 
 
 
 
 
 
 
 
 
 
 
 
101
  # Combining Flash Diffusion with Existing ControlNets 🎨
102
 
103
  FlashSDXL can also be combined with existing ControlNets to unlock few steps generation in a **training free** manner. It can be integrated straight to Hugging Face pipelines. See an example below.
 
98
  <img style="width:400px;" src="images/corgi.jpg">
99
  </p>
100
 
101
+ # How to use in Comfy?
102
+
103
+ To use FlashSDXL locally using Comfyui you need to :
104
+
105
+ 1. Make sure your comfyUI install is up to date
106
+ 2. Download the checkpoint from [huggingface](https://huggingface.co/jasperai/flash-sdxl).
107
+ In case you wonder how, go to "Files and Version" go to `comfy/` folder and hit the download button next to the `FlashSDXL.safetensors`
108
+ 3. Move the new checkpoint file to your local `comfyUI/models/loras/.` folder
109
+ 4. Use it as a LoRA on top of `sd_xl_base_1.0_0.9vae.safetensors`, a simple comfyui `workflow.json` is provided in this repo (available in the same `comfy/` folder)
110
+
111
+ > Disclaimer : Model has been trained to work with a cfg scale of 1 and a lcm scheduler but parameters can be tweaked a bit.
112
+
113
  # Combining Flash Diffusion with Existing ControlNets 🎨
114
 
115
  FlashSDXL can also be combined with existing ControlNets to unlock few steps generation in a **training free** manner. It can be integrated straight to Hugging Face pipelines. See an example below.