# PixArt-alpha/PixArt-alpha

PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis

Repository: https://github.com/PixArt-alpha/PixArt-alpha
Canonical: https://ross.abutalabs.com/products/pixart-alpha
Homepage: https://pixart-alpha.github.io/
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-10-31T01:20:53+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 75
- inputs: {"age_days": 1056, "days_push": 672, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3304, forks 202 (observed 2026-08-28T04:07:55.776871+00:00)

## What it is
PixArt-α is a Transformer-based text-to-image diffusion model with PyTorch model definitions, pre-trained weights, and inference/training code. It achieves photorealistic 1024px image synthesis competitive with SDXL and Midjourney at a fraction of the usual training cost.

## Use cases
- generate photorealistic images from text prompts
- train a text-to-image diffusion model cheaply
- run a 1024px T2I model locally with PyTorch
- integrate a diffusion transformer into diffusers pipelines
- research fast training of diffusion models

## When to choose
- you need high-quality text-to-image generation with open weights
- you want a low-cost, reproducible training recipe for T2I diffusion models
- you want a model already supported in Hugging Face Diffusers and ComfyUI

## When to avoid
- you need video or audio generation
- you want a polished end-user app rather than research code
- you need the latest model iteration - the successor PixArt-Σ may be more current

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, llm-training
- domain: deep-learning, image-processing, artificial-intelligence
- platform: python
- tags: text-to-image, diffusion-transformer, diffusers, pytorch, generative-ai, research-code, gpu, linux

## Member repositories
- PixArt-alpha/PixArt-alpha (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:55.776871+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:42:39.647821+00:00, confidence not recorded.
  - readme: https://github.com/PixArt-alpha/PixArt-alpha (fetched 2026-08-28T04:07:55.776871+00:00, sha e3e5ef8daa17)
  - homepage: https://pixart-alpha.github.io/ (fetched 2026-08-29T09:35:24.696046+00:00, sha e8fe5650ea2a)
- Data as of 2026-08-30T08:39:29.467469+00:00.