ermongroup/ddim
A PyTorch implementation of Denoising Diffusion Implicit Models for faster image generation via accelerated diffusion sampling.
Not currently ranked — collecting fresh signals.
star history
Implements DDIM, a sampling method for implicit models trained similarly to DDPM but with significantly reduced sampling time and compute. The technique provides deterministic and faster reverse diffusion processes compared to standard DDPM, enabling efficient image generation. It integrates with the Hugging Face Diffusers library and can be combined with larger models like Stable Diffusion.
Frequently asked
- What is ermongroup/ddim?
- A PyTorch implementation of Denoising Diffusion Implicit Models for faster image generation via accelerated diffusion sampling.
- Is ddim open source?
- Yes — ermongroup/ddim is open source, released under the MIT license.
- What language is ddim written in?
- ermongroup/ddim is primarily written in Python.
- How popular is ddim?
- ermongroup/ddim has 1.8k stars on GitHub.
- Where can I find ddim?
- ermongroup/ddim is on GitHub at https://github.com/ermongroup/ddim.