Release MetaEmbed model checkpoints on Hugging Face

Hi @MrZilinXiao 🤗

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/2509.18095.
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.

I saw that you've released the code for the MetaEmbed framework on GitHub, which is fantastic! I noticed you achieved state-of-the-art results on MMEB and ViDoRe with models scaling up to 32B.

Would you like to host the pre-trained MetaEmbed model checkpoints (e.g., the Qwen2.5 and Llama 3.2 Vision variants) on https://huggingface.co/models?

Hosting on Hugging Face will give your work more visibility and enable better discoverability. We can add specific metadata tags (like `visual-document-retrieval`) in the model cards so that people find the models easier, and directly link them to the paper page.

If you're down, leaving a guide [here](https://huggingface.co/docs/hub/models-uploading). Since your implementation leverages `transformers`, users would be able to load your models seamlessly.

After uploaded, we can also link the models to the paper page (read [here](https://huggingface.co/docs/hub/en/model-cards#linking-a-paper)) so people can discover your work.

You can also build a demo for your model on [Spaces](https://huggingface.co/spaces), we can provide you a ZeroGPU [grant](https://huggingface.co/docs/hub/en/spaces-gpus#community-gpu-grants), which gives you A100 GPUs for free.

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

Kind regards,

Niels