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: Access to facebook/sam3 Rejected
Hello SAM 3 Team,
My HF username is: yyyyh1. My access request to facebook/sam3 was rejected, and I would like to kindly ask for a manual review, or for my request status to be reset to pending so I can resubmit.
Research purpose:
I am conducting non-commercial academic research on biomedical image analysis, specifically the segmentation of vasculature in mouse imaging data. I plan to use SAM 3 for promptable segmentation of blood vessels in microscopy/photographic images, and to evaluate its zero-shot performance against other segmentation approaches (e.g.,MedSAM) for this task.
The checkpoint will be used solely for research and offline evaluation on my local workstation. There will be no commercial use, no clinical application, and no redistribution of the model weights.
I will fully comply with the SAM License and provide appropriate attribution in any resulting publications.
Thank you very much for your time and consideration.
Best regards