mit-han-lab/lpd
[ICLR 2026 Oral] Locality-aware Parallel Decoding for Efficient Autoregressive Image Generation
This tool helps researchers and AI practitioners generate high-quality images from scratch significantly faster than traditional methods. It takes a model (already trained or one you train) and outputs images, such as those generated from class-conditional inputs or text prompts. You would use this if you need to rapidly create many images for datasets, prototyping, or research experiments, without compromising visual quality.
Use this if you are a researcher or practitioner in generative AI who needs to generate high-fidelity images quickly and efficiently, especially for tasks like dataset expansion or model evaluation.
Not ideal if you are looking for an off-the-shelf image editing tool or a simple API for general image creation without deep dives into model parameters.
Stars
91
Forks
7
Language
Python
License
MIT
Category
Last pushed
Mar 12, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/mit-han-lab/lpd"
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...