31 lines
1.3 KiB
Markdown
31 lines
1.3 KiB
Markdown
# ComfyUI-RetroDiffusion-API-Node
|
|
|
|
A ComfyUI custom node for the [Retro Diffusion API](https://retrodiffusion.ai/)
|
|
|
|
<img width="795" height="541" alt="comfyui_retrodiffusion_api_node showcase" src="https://github.com/user-attachments/assets/9f2c9634-74e6-4b61-a8f2-6e08132f2f4c" />
|
|
|
|
## How to Install
|
|
1. Go to [Retro Diffusion](https://retrodiffusion.ai/) and sign up for an account if you don't have one.
|
|
2. Create an API key from the developer tools.
|
|
3. In the root folder of your ComfyUI installation, create a `.env` file if it doesn't exist already.
|
|
4. Add the following to your `.env`:
|
|
```
|
|
RD_API_KEY=your-key
|
|
```
|
|
5. Clone this repository into your `ComfyUI/custom_nodes` folder.
|
|
6. Start ComfyUI.
|
|
|
|
Now you can create a `Retro Diffusion API Node` in your workflows.
|
|
|
|
## Features
|
|
- Text-to-image and image-to-image
|
|
- Outputs credit cost of the image and your remaining credits too
|
|
- Supports `RD_Fast`, `RD_PLUS`, and `Animation` styles as of the latest commit
|
|
|
|
More info on the Retro Diffusion API itself can be found here - https://github.com/Retro-Diffusion/api-examples
|
|
|
|
## Notes
|
|
- `RD_PLUS` and `Animation` styles are more expensive to generate than `RD_Fast`.
|
|
- Check the console if you get any errors, and feel free to create an issue.
|
|
- There is an upload size limit for image-to-image, so keep that in mind (don't send large images as input).
|