TL; DR: PixArt-α is a Transformer-based T2I diffusion model whose image generation quality is competitive with state-of-the-art image generators (e.g., Imagen, SDXL, and even Midjourney), …

2023年9月30日 · This paper introduces PIXART-$\alpha$, a Transformer-based T2I diffusion model whose image generation quality is competitive with state-of-the-art image generators …

2023年10月6日 · 华为发布了新的文生图模型 PixArt-α，它主打的是低训练成本。PixArt-α采用Meta之前提出的DiT架构，模型参数只有0.6B，它的训练时长只有SD 1.5的10.8%，而且只使 …

This paper introduces PIXART-α, a Transformer-based T2I diffusion model whose image generation quality is competitive with state-of-the-art image generators (e.g., Imagen, SDXL, …

The chart above evaluates user preference for Pixart-α over SDXL 0.9, Stable Diffusion 2, DALLE-2 and DeepFloyd. The Pixart-α base model performs comparable or even better than …

2024年1月16日 · TL;DR: PixArt-α is a cost-effective, eco-friendly text-to-image diffusion model that delivers high-quality images, reduces training expenses and CO2 emissions, and fosters …

2024年1月10日 · This collection organize all the PixArt-Alpha related models, datasets and so on.

PixArt Alpha is a Transformer-based T2I diffusion model that generates photorealistic images from text. The model’s image generation quality is competitive with state-of-the-art image …

2024年3月18日 · 本文介绍了 PiXART-α\alphaα，这是一种基于 Transformer 的 T2I 扩散模型，其图像生成质量可与最先进的图像生成器（例如，Imagen、SDXL，甚至 Midjourney）竞争，达 …

This collection organize all the PixArt-Alpha related models, datasets and so on. • 9 items • Updated May 4 • 2