PixArt-alpha/PixArt-sigma
A diffusion transformer model that generates high-resolution 4K images from text prompts.
Not currently ranked — collecting fresh signals.
star history
PixArt-Σ is a text-to-image generation model based on a diffusion transformer architecture. It uses a weak-to-strong training methodology to progressively improve model capability. The repository provides PyTorch model definitions, pre-trained weights, and inference/sampling code, enabling generation of 4K-resolution images from text descriptions.
Frequently asked
- What is PixArt-alpha/PixArt-sigma?
- A diffusion transformer model that generates high-resolution 4K images from text prompts.
- Is PixArt-sigma open source?
- Yes — PixArt-alpha/PixArt-sigma is open source, released under the Apache-2.0 license.
- What language is PixArt-sigma written in?
- PixArt-alpha/PixArt-sigma is primarily written in Python.
- How popular is PixArt-sigma?
- PixArt-alpha/PixArt-sigma has 1.9k stars on GitHub.
- Where can I find PixArt-sigma?
- PixArt-alpha/PixArt-sigma is on GitHub at https://github.com/PixArt-alpha/PixArt-sigma.