Block Diffusion – Generate images from text prompts
A new era of content creation is here. With machine learning and advanced models, we are now capable of creating unique images based on text prompts alone. We are only lmiited by our imagination. Block Diffusion lets you use artifical intelligence to unlock your creativity and give unique experiences to your visitors.
This plugin interfaces with the Replicate API and allows you to run open-source models via their cloud API. Add the block to a page and select a model to enter your API token.
Semi interactive demo at https://www.block-diffusion.com/
Features
- Stable Diffusion – A latent text-to-image diffusion model capable of generating photo-realistic images given any text input
- Stable Diffusion V2 test version
- Text to Pokémon – Generate Pokémon from a text description
- Open Journey – Stable Diffusion fine tuned on Midjourney v4 images.
- Multiple outputs
- Image input with basic controls
- Download, copy, or import into the editor
- Prompt suggestion + preview
Pro version
All pro features will be added to this open source version on wordpress.org. You may sponsor the project on GitHub (there are monthly and one-time payment options). Thank you.
Upcoming Features
- Add an image mask to only generate specific areas
- View your prediction history
More Info
- Follow @kevinbatdorf on Twitter
- View on GitHub
Terms and Privacy
This plugin provides an interface to the Replicate API and requires you have an active account at replicate.com with an active API token. This plugin is not affiliated with Replicate and only uses their public Http API. Your servers speak directly with replicate.com. – Replicate terms – Replicate privacy policy
We also offer additional features on top of the Replicate API, such as prompt/output sharing (opt in) and prompt inspiration. – Block Diffusion terms – Block Diffusion privacy policy