V
VAR
FoundationVision/VAR
[NeurIPS 2024 Best Paper Award][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!
★8.7kstars
Jupyter Notebook
MIT
Updated: Today
📋 Project at a Glance
Tap to expand
What's this?A auto-regressive-model/autoregressive-models tool in the Models category, built with Jupyter Notebook, open-source
Who made it?Maintained by FoundationVision team, 8.7K⭐ on GitHub, #174 out of 3201 in Models
Why does it exist?In the Models space, auto-regressive-model workflows faced efficiency bottlenecks. VAR was built by FoundationVision to address these autoregressive-models challenges.
What can it do?Key use cases: diffusion-models, generative-ai, generative-model
How to install with AI?Use an AI coding assistant to follow the README and automatically handle the install and environment setup.
🔗 github.com/FoundationVision/VAR
🔗 github.com/FoundationVision/VAR
Topics
auto-regressive-modelautoregressive-modelsdiffusion-modelsgenerative-aigenerative-modelgptgpt-2image-generationlarge-language-modelsneuripstransformersvision-transformer