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 Review or Reset of Rejected SAM 3 Access β Academic Robotics Real2Sim Research
Dear SAM 3 Team,
My access request for SAM 3 was rejected, and I would appreciate a review of the decision.
My Hugging Face username is ShibaFw and I'm a student from Shanghai Jiao Tong University. I am conducting non-commercial academic research on real-to-simulation (real2sim) for robotics. I plan to use SAM 3 to segment and track objects in real-world images and videos, providing masks for downstream 3D reconstruction and the creation of robotic manipulation scenes in simulation. This work involves studying and reproducing research pipelines such as Do as I Do and SimFoundry.
Could you please reconsider my application or clarify whether any information needs correction? If necessary, I would appreciate a reset of my rejected request so that I can resubmit accurate and complete information. I am happy to provide additional details through an appropriate private channel.
I will comply with the applicable model license and access requirements and will not redistribute the model weights.
Thank you for your time and consideration.
Best regards,
Kehang Chai