NVlabs
Release SANA-Video on Hugging Face
Hi @lawrence-cj 🤗
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/2509.24695. 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.
Your paper introduces "SANA-Video", an exciting new diffusion model for efficient video generation. I saw on your project page (https://nvlabs.github.io/Sana/Video) that the code and model will be publicly released, and the GitHub repo also lists "Video Generation" as a future task.
Would you like to host the SANA-Video model you've pre-trained on https://huggingface.co/models once it's ready? Hosting on Hugging Face will give you more visibility/enable better discoverability. We can add tags in the model cards so that people find the models easier, link it 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.
After uploaded, we can also link the models to the paper page (read here) so people can discover your model.
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 once the model is released :)
Kind regards,
Niels
Hey, could you please post a message here when it's release (so I can get a notification haha)
Thanks for you guys' attention.
@NielsRogge , Hi Niels, thanks so much for your notice and guidance here. I'm happy to support SANA-Video on the Huinggafce platform on day zero, as we previously always did. I'll ping you guys when we finish the preparation. It'll come very soon!
I updated my diffusers, but I still could not use SanaVideo. See the log below. Could you tell me how to solve this problem? thx a lot
pip install git+https://github.com/huggingface/diffusers
Collecting git+https://github.com/huggingface/diffusers
Cloning https://github.com/huggingface/diffusers to /tmp/pip-req-build-lyrqi4lu
Running command git clone --filter=blob:none --quiet https://github.com/huggingface/diffusers /tmp/pip-req-build-lyrqi4lu
Resolved https://github.com/huggingface/diffusers to commit 8f80dda193f79af3ccd0f985906d61123d69df08
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... done
Requirement already satisfied: importlib_metadata in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (8.5.0)
Requirement already satisfied: filelock in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (3.16.1)
Requirement already satisfied: httpx<1.0.0 in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (0.28.1)
Requirement already satisfied: huggingface-hub<2.0,>=0.34.0 in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (0.36.0)
Requirement already satisfied: numpy in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (1.26.4)
Requirement already satisfied: regex!=2019.12.17 in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (2024.11.6)
Requirement already satisfied: requests in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (2.32.3)
Requirement already satisfied: safetensors>=0.3.1 in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (0.4.5)
Requirement already satisfied: Pillow in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (10.4.0)
Requirement already satisfied: anyio in /opt/venv/lib/python3.10/site-packages (from httpx<1.0.0->diffusers==0.36.0.dev0) (4.7.0)
Requirement already satisfied: certifi in /opt/venv/lib/python3.10/site-packages (from httpx<1.0.0->diffusers==0.36.0.dev0) (2024.8.30)
Requirement already satisfied: httpcore==1.* in /opt/venv/lib/python3.10/site-packages (from httpx<1.0.0->diffusers==0.36.0.dev0) (1.0.7)
Requirement already satisfied: idna in /opt/venv/lib/python3.10/site-packages (from httpx<1.0.0->diffusers==0.36.0.dev0) (3.10)
Requirement already satisfied: h11<0.15,>=0.13 in /opt/venv/lib/python3.10/site-packages (from httpcore==1.*->httpx<1.0.0->diffusers==0.36.0.dev0) (0.14.0)
Requirement already satisfied: fsspec>=2023.5.0 in /opt/venv/lib/python3.10/site-packages (from huggingface-hub<2.0,>=0.34.0->diffusers==0.36.0.dev0) (2024.10.0)
Requirement already satisfied: packaging>=20.9 in /opt/venv/lib/python3.10/site-packages (from huggingface-hub<2.0,>=0.34.0->diffusers==0.36.0.dev0) (24.2)
Requirement already satisfied: pyyaml>=5.1 in /opt/venv/lib/python3.10/site-packages (from huggingface-hub<2.0,>=0.34.0->diffusers==0.36.0.dev0) (6.0.2)
Requirement already satisfied: tqdm>=4.42.1 in /opt/venv/lib/python3.10/site-packages (from huggingface-hub<2.0,>=0.34.0->diffusers==0.36.0.dev0) (4.67.1)
Requirement already satisfied: typing-extensions>=3.7.4.3 in /opt/venv/lib/python3.10/site-packages (from huggingface-hub<2.0,>=0.34.0->diffusers==0.36.0.dev0) (4.15.0)
Requirement already satisfied: hf-xet<2.0.0,>=1.1.3 in /opt/venv/lib/python3.10/site-packages (from huggingface-hub<2.0,>=0.34.0->diffusers==0.36.0.dev0) (1.2.0)
Requirement already satisfied: exceptiongroup>=1.0.2 in /opt/venv/lib/python3.10/site-packages (from anyio->httpx<1.0.0->diffusers==0.36.0.dev0) (1.2.2)
Requirement already satisfied: sniffio>=1.1 in /opt/venv/lib/python3.10/site-packages (from anyio->httpx<1.0.0->diffusers==0.36.0.dev0) (1.3.1)
Requirement already satisfied: zipp>=3.20 in /opt/venv/lib/python3.10/site-packages (from importlib_metadata->diffusers==0.36.0.dev0) (3.21.0)
Requirement already satisfied: charset-normalizer<4,>=2 in /opt/venv/lib/python3.10/site-packages (from requests->diffusers==0.36.0.dev0) (3.4.0)
Requirement already satisfied: urllib3<3,>=1.21.1 in /opt/venv/lib/python3.10/site-packages (from requests->diffusers==0.36.0.dev0) (1.26.20)
colligo@pluto-prod-zongzew3-anh68f-7-0:/mnt/localssd/CausVid-internal$ python
Python 3.10.16 (main, Dec 4 2024, 08:53:37) [GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from diffusers import SanaVideoPipeline
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ImportError: cannot import name 'SanaVideoPipeline' from 'diffusers' (/opt/venv/lib/python3.10/site-packages/diffusers/__init__.py)
>>>
I updated my diffusers, but I still could not use SanaVideo. See the log below. Could you tell me how to solve this problem? thx a lot
pip install git+https://github.com/huggingface/diffusers Collecting git+https://github.com/huggingface/diffusers Cloning https://github.com/huggingface/diffusers to /tmp/pip-req-build-lyrqi4lu Running command git clone --filter=blob:none --quiet https://github.com/huggingface/diffusers /tmp/pip-req-build-lyrqi4lu Resolved https://github.com/huggingface/diffusers to commit 8f80dda193f79af3ccd0f985906d61123d69df08 Installing build dependencies ... done Getting requirements to build wheel ... done Preparing metadata (pyproject.toml) ... done Requirement already satisfied: importlib_metadata in /opt/venv/lib/python3.10/site-packages (from diffusers==0.36.0.dev0) (8.5.0)
run pip install -U git+https://github.com/huggingface/diffusers to update pip package
Also this is in current development, I've found the development fork here https://github.com/lawrence-cj/diffusers/tree/feat/sana-video
Thanks a lot for your reply! I’ve got it working now.
However, I noticed that the current VAE is the 8× compression AutoencoderKLWan, rather than the 32× compression DCAE mentioned in the paper. Do you have any plans to release the model with the DCAE version?