Release GGUF models on Hugging Face

[NielsRogge]

Hi @liborui 🤗

I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work on Arxiv and was wondering whether you would like to submit it to hf.co/papers to improve its discoverability. If you are one of the authors, you can submit it at https://huggingface.co/papers/submit.

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, and add Github and project page URLs.

I noticed in your Github README that you have a "Hugging Face coming soon" badge. Would you like to host the converted GGUF model checkpoints (like pi0.5, HY-VLA, and LingBot-VA) on https://huggingface.co/models under an official organization?

Hosting on Hugging Face will give your project more visibility and enable better discoverability. We can add tags in the model cards so that people find the models easier, link them to the paper page, etc.

If you're down, leaving a guide here. If you want to directly upload models through the UI or via CLI, people can easily use hf_hub_download to fetch them.

After uploaded, we can also link the models to the paper page (read here) so people can discover your work.

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

Kind regards,

Niels

[liborui]

Hi @NielsRogge , thanks a lot for reaching out and for the suggestion!

Yes, we’re definitely interested. We have now uploaded the converted GGUF checkpoints to our official Hugging Face organization here:

https://huggingface.co/SEU-PAISys/Embodied.cpp

This repo currently hosts the GGUF weights we prepared for Embodied.cpp, including:

• pi0.5
• HY-VLA (in few days)
• LingBot-VA (in few days)

We also agree that linking the paper page, model artifacts, GitHub repo, and project page on Hugging Face would be very helpful for discoverability.

Regarding the paper submission flow: when I tried to submit it, I ran into the restriction shown in the screenshot, namely that only authors with at least one linked paper on HF can submit to Daily Papers yet. I have already submitted the paper-link / claiming request on Hugging Face, and I’m currently waiting for it to be processed. Once that is approved, I’ll complete the paper submission and link the related artifacts accordingly.

Thanks again for the helpful guidance and support!

Best,
Borui

[NielsRogge]

Hi @liborui, thanks for uploading the GGUF checkpoints to the Hugging Face hub and for the detailed update. I can see that the pi0.5 and HY-VLA GGUF weights are already available in the repo.

A quick note on the linking: your model repo (SEU-PAISys/Embodied.cpp) is already linked to the paper page at https://huggingface.co/papers/2607.02501. Because the model card README mentions the arXiv abstract URL (https://arxiv.org/abs/2607.02501), the artifact shows up automatically on the paper page, so that part is already taken care of and you don't need to wait for it. The GitHub README also already references the Hugging Face repo (the badge and the "Get GGUF Weights" section), so everything is well connected there too.

Regarding claiming the paper: once your Hugging Face account claiming request is approved, you can claim the paper page by clicking on your name on the paper page, which will also display it on your public profile.

Looking forward to the LingBot-VA GGUF release in the coming days. Feel free to ping here once it's uploaded, or let me know if you run into any issues. Thanks again for making the work available on the hub.

[vimalnakrani08]

PR #2 adds the runnable text-generation path that a GGUF release needs. Quantized VLM conversions (q4_K/q5_K/q6_K/q8_0, dequant-verified per tensor against the safetensors) are ready to publish once it lands.