dome272/Paella
A text-to-image diffusion model that generates high-fidelity images in fewer than 10 sampling steps.
Not currently ranked — collecting fresh signals.
star history
Paella is a text-to-image diffusion model that generates high-fidelity images in under 10 sampling steps. It operates on a compressed and quantized latent space, conditions on CLIP embeddings, and achieves fast inference under 500ms per image. The model supports text-conditional generation, latent interpolation, and image manipulation tasks including inpainting, outpainting, and structural editing.
Frequently asked
- What is dome272/Paella?
- A text-to-image diffusion model that generates high-fidelity images in fewer than 10 sampling steps.
- Is Paella open source?
- Yes — dome272/Paella is open source, released under the MIT license.
- What language is Paella written in?
- dome272/Paella is primarily written in Jupyter Notebook.
- How popular is Paella?
- dome272/Paella has 748 stars on GitHub.
- Where can I find Paella?
- dome272/Paella is on GitHub at https://github.com/dome272/Paella.