Happy 1st anniversary 🤗 Diffusers!

🤗 Diffusers is happy to celebrate its first anniversary! It has been an exciting year, and we're proud and grateful for how far we've come thanks to our community and open-source contributors. Last year, text-to-image models like DALL-E 2, Imagen, and Stable Diffusion captured the world's attention with their ability to generate stunningly photorealistic images from text, sparking a massive surge of interest and development in generative AI. But access to these powerful models was limited.

At Hugging Face, our mission is to democratize good machine learning by collaborating and helping each other build an open and ethical AI future together. Our mission motivated us to create the 🤗 Diffusers library so everyone can experiment, research, or simply play with text-to-image models. That’s why we designed the library as a modular toolbox, so you can customize a diffusion model’s components or just start using it out-of-the-box.

As 🤗 Diffusers turns 1, here’s an overview of some of the most notable features we’ve added to the library with the help of our community. We are proud and immensely grateful for being part of an engaged community that promotes accessible usage, pushes diffusion models beyond just text-to-image generation, and is an all-around inspiration.

Table of Contents

• Striving for photorealism
• Video pipelines
• Text-to-3D models
• Image editing pipelines
• Faster diffusion models
• Ethics and safety
• Support for LoRA
• Torch 2.0 optimizations
• Community highlights
• Building products with 🤗 Diffusers
• Looking forward

Striving for photorealism

Generative AI models are known for creating photorealistic images, but if you look closely, you may notice certain things that don't look right, like generating extra fingers on a hand. This year, the DeepFloyd IF and Stability AI SDXL models made a splash by improving the quality of generated images to be even more photorealistic.

Video pipelines

Text-to-3D models

Try it out today with the ShapEPipeline and ShapEImg2ImgPipeline.

Simo

Qing