CVL-UESTC/MVAR
ICLR 2026-MVAR: Visual Autoregressive Modeling with Scale and Spatial Markovian Conditioning
This project helps machine learning engineers and researchers generate high-quality images more efficiently. It takes existing pre-trained visual autoregressive models and optimizes their inference process. The output is generated images with significantly reduced GPU memory consumption, making complex image generation tasks more accessible and faster.
Use this if you are a machine learning researcher or engineer working on large-scale image generation and need to reduce the GPU memory footprint during inference.
Not ideal if you are looking for an off-the-shelf image generation tool without diving into model training and optimization.
Stars
36
Forks
1
Language
Python
License
MIT
Category
Last pushed
Mar 13, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/CVL-UESTC/MVAR"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
hao-ai-lab/FastVideo
A unified inference and post-training framework for accelerated video generation.
ModelTC/LightX2V
Light Image Video Generation Inference Framework
thu-ml/TurboDiffusion
TurboDiffusion: 100–200× Acceleration for Video Diffusion Models
PKU-YuanGroup/Helios
Helios: Real Real-Time Long Video Generation Model
PKU-YuanGroup/MagicTime
[TPAMI 2025🔥] MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators