horseee/DeepCache
DeepCache is a caching-based technique that accelerates diffusion model inference without training by reusing intermediate features across denoising steps.
Not currently ranked — collecting fresh signals.
star history
DeepCache is a CVPR 2024 research project that accelerates diffusion models during inference by caching and reusing intermediate features across denoising steps. The method achieves up to 2.3x speedup on Stable Diffusion with minimal quality loss. It supports Stable Diffusion v1.5, SDXL, Stable Video Diffusion, and various pipelines including inpainting and img2img. The technique is training-free and integrates into the HuggingFace Diffusers library as a plug-and-play optimization.
Frequently asked
- What is horseee/DeepCache?
- DeepCache is a caching-based technique that accelerates diffusion model inference without training by reusing intermediate features across denoising steps.
- Is DeepCache open source?
- Yes — horseee/DeepCache is open source, released under the Apache-2.0 license.
- What language is DeepCache written in?
- horseee/DeepCache is primarily written in Python.
- How popular is DeepCache?
- horseee/DeepCache has 970 stars on GitHub.
- Where can I find DeepCache?
- horseee/DeepCache is on GitHub at https://github.com/horseee/DeepCache.