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")# pip install -U transformers accelerate # 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 reconsideration of rejected SAM 3 access
#697
by yuxin-77 - opened
Dear Meta team,
I am working on reproducing HexaShell, a research project focused on robotic grasping. This work relies on the official SAM 3 model to perform visual segmentation and scene understanding for grasping tasks.
My previous access application was rejected, and the form no longer allows me to resubmit. I would like to request another review, or a reset of my request status so that I can reapply with corrected information.
I will use SAM 3 strictly for academic research and paper reproduction. All inference will run locally on our own hardware. I will not use the model for any commercial purposes, will not redistribute the model weights, and will fully comply with the license terms and Meta Privacy Policy.
Could you please reconsider my application and grant me access to SAM 3? Having access to the model would allow me to complete my reproduction experiments.
Thank you very much for your time.
Best regards,
yuxin-77