lmxyy/sige
[NeurIPS 2022, T-PAMI 2023] Efficient Spatially Sparse Inference for Conditional GANs and Diffusion Models
Implements tiling-based sparse convolution that selectively computes only edited regions by gathering active blocks, reducing MACs by 7-18x while preserving quality. Supports conditional GANs (GauGAN) and diffusion models (DDPM, DDIM, Stable Diffusion) with multi-backend inference on NVIDIA GPUs, Apple M1/MPS, and CPU via PyTorch. Composable with model compression techniques like GAN Compression for cumulative speedups up to 47x on specialized hardware.
268 stars. No commits in the last 6 months. Available on PyPI.
Stars
268
Forks
13
Language
Python
License
—
Category
Last pushed
Mar 18, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/lmxyy/sige"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
UCSC-VLAA/story-iter
[ICLR 2026] A Training-free Iterative Framework for Long Story Visualization
PaddlePaddle/PaddleMIX
Paddle Multimodal Integration and eXploration, supporting mainstream multi-modal tasks,...
keivalya/mini-vla
a minimal, beginner-friendly VLA to show how robot policies can fuse images, text, and states to...
adobe-research/custom-diffusion
Custom Diffusion: Multi-Concept Customization of Text-to-Image Diffusion (CVPR 2023)
byliutao/1Prompt1Story
🔥ICLR 2025 (Spotlight) One-Prompt-One-Story: Free-Lunch Consistent Text-to-Image Generation...