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 Access to SAM 3 Model Weights for Academic Research
Dear SAM 3 Team,
I am a PhD student at Hangzhou Dianzi University, China, currently conducting research in computer vision, visual counting, and intelligent video analysis.
I am writing to request access to the facebook/sam3 repository on Hugging Face, particularly the model weights such as sam3.pt. My research focuses on exploring how SAM 3's promptable segmentation capabilities can support open-category object counting and improve object localization in complex visual scenes.
I intend to use the model strictly for academic and non-commercial research purposes and will comply with the applicable license terms and usage policies.
Thank you for your time and consideration.
Best regards,
Shuo Zhang
PhD Student
Hangzhou Dianzi University, China