Text-to-Image Diffusion Models

11 repos

Model implementations and optimizations for generating images from text descriptions using diffusion-based architectures. This cluster centers on production-ready implementations of models like Sana and PixArt, formatted as Hugging Face diffusers-compatible checkpoints with safetensors serialization. Repositories here focus on scaling these generative models across different resolutions and parameter counts, making them accessible for inference and fine-tuning workflows.

Related clusters