Apply for a GPU community grant: Academic project
Pixel Perfect MegaMed is a vision–language diffusion model for ultra-high-resolution chest X-ray synthesis (1024×1024 and up to 2048×2048). The model enables semantically controlled medical image generation conditioned on clinical findings such as pneumothorax, edema, and pleural effusion.
This demo accompanies our research paper and allows users to generate high-resolution medical images interactively using Stable Diffusion XL with LoRA fine-tuning. GPU resources are required because the model relies on SDXL and interactive inference requires GPU acceleration.
A member of the Hugging Face open-source team suggested hosting the model and demo on Hugging Face to improve discoverability, so we created this Space to make the model accessible to the research community.
Granting GPU access would allow researchers and practitioners to interactively explore the model and better understand the capabilities of high-resolution medical image generation.
Links:
Paper: https://arxiv.org/abs/2507.12698
Project page: https://tehraninasab.github.io/pixelperfect-megamed/
Code: https://github.com/tehraninasab/pixelperfect-megamed
Model weights: https://huggingface.co/tehraninasab/pixelperfect-megamed-lora
Hi @tehraninasab , we've assigned ZeroGPU to this Space. Please check the compatibility and usage sections of this page so your Space can run on ZeroGPU.
If you can, we ask that you upgrade to Pro ($9/month) to enjoy higher ZeroGPU quota and other features like Dev Mode, Private Storage, and more: hf.co/pro