Release 3D MedDiffusion checkpoints on Hugging Face

[NielsRogge]

Hi @Mors20 🤗

I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work through Hugging Face's daily papers as yours got featured: https://huggingface.co/papers/2412.13059.
The paper page lets people discuss about your paper and lets them find artifacts about it (your models for instance),
you can also claim the paper as yours which will show up on your public profile at HF, add Github and project page URLs.

It's fantastic to see that the MMorss/3D-MedDiffusion model is already hosted on huggingface.co/models! I also noticed in your GitHub README that you provide additional pre-trained checkpoints for 8x and 4x downsampling via a Google Drive link.

Would you be interested in hosting these additional 3D MedDiffusion checkpoints directly on https://huggingface.co/models as well? This would consolidate all your model artifacts in one place, enabling even better visibility and discoverability for your work. We can add tags in the model cards so that people can find and use these models more easily, link them all to the paper page, etc.

If you're down, leaving a guide here. If it's a custom PyTorch model, you can use the PyTorchModelHubMixin
class which adds from_pretrained and push_to_hub to the model which lets you to upload the model and people to download and use models right away.
If you do not want this and directly want to upload model through UI or however you want, people can also use hf_hub_download.

We encourage researchers to push each distinct model checkpoint (e.g., the 4x and 8x downsampling versions) to a separate model repository (or clearly separated subfolders within the same repo), so that things like download stats and versioning also work effectively. After uploaded, we can also link all these checkpoints to the paper page (read here) so people can discover all your work.

You can also build a demo for your model on Spaces, we can provide you a ZeroGPU grant, which gives you A100 GPUs for free.

Let me know if you're interested/need any guidance :)

Kind regards,

Niels