Instructions to use facebook/sam3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/sam3 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("mask-generation", model="facebook/sam3")# Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("facebook/sam3") model = AutoModel.from_pretrained("facebook/sam3", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Reset of Rejected SAM3 Access
Hello Meta SAM3 Team,
My access request to facebook/sam3 was rejected. Could you please manually reconsider my request, or reset its status so that I can submit corrected and more complete information?
Hugging Face username: zihan wu
Affiliation: Xuanya Robotics
Role: Robotics Research and Development
Intended use:
I plan to use SAM3 locally for person segmentation in synchronized multi-camera recordings. The resulting masks will be propagated across video frames and used to prepare person-free scenes for 4D Gaussian Splatting reconstruction.
This work is for internal research and prototyping at Xuanya Robotics. The model will run locally. I will comply with the applicable license and usage requirements, and I will not redistribute the model weights or share my Hugging Face access token.
If my original submission was incomplete or unclear, I would be happy to provide any additional information required for review.
Thank you for your time and consideration.
Best regards,
Zihan Wu
Xuanya Robotics