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
Request for Manual Review / Reset of Rejected SAM 3 Access
Hello Meta SAM 3 Team,
My access request for facebook/sam3 was recently rejected, likely because some of the information in my original application was incomplete.
I have now updated and completed my Hugging Face profile. Could you please manually review my request again, or reset the rejected status so that I can submit the access form with the corrected information?
Hugging Face username: guoyang1
Affiliation: The Hong Kong University of Science and Technology (Guangzhou)*
Role: PhD Student in Robotics and Autonomous Systems*
Intended use:
I plan to use SAM 3 solely for non-commercial academic research in robotic perception and embodied AI. Specifically, I would like to evaluate it for object segmentation and tracking in robot and egocentric videos, and potentially integrate the segmentation results into robot-learning data-processing pipelines.
I will comply with the applicable model license, Meta policies, and Hugging Face terms. I will not redistribute the model weights or use the model for prohibited, harmful, surveillance, military, or biometric applications.
Thank you for considering my request.
Best regards,
Guoyang Zhao