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 to reset rejected SAM3 access request
Hello SAM3 Team:
My access request to facebook/sam3.1 was rejected, and I would like to kindly request a manual review or a reset of the request status so I can resubmit with complete information.
I am a graduate student conducting non-commercial academic research on robotic perception.
My intended use of SAM3 is to evaluate and fine-tune promptable video segmentation and tracking models on indoor robotics datasets. Specifically, I plan to use SAM 3.1 to generate and track object masks in prerecorded robot videos, and study temporal consistency under camera motion and occlusion.
The model will run locally on our research workstation. I will not redistribute or republish the model weights, nor provide public access to the checkpoint. I will comply with the SAM license and all applicable usage restrictions.
Hugging Face username: hejianda
Role: Graduate Student
Intended use: Non-commercial academic research in robotic perception and video object tracking.
If possible, could you please manually review my request or reset it to pending so I can reapply?
Thank you for your time and consideration.