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 Manual Review
Dear Meta SAM 3 Team,
My access request for facebook/sam3 on Hugging Face was recently rejected, and I am currently unable to submit a new application.
I am affiliated with Peking University and work on robotics and embodied AI research.
I would like to use SAM 3 for non-commercial academic research, particularly in:
Visual perception and object segmentation for robotic manipulation.
Segmenting task-relevant objects and interactive regions in indoor environments.
Evaluating prompt-based image and video segmentation for embodied AI applications.
The model will be used locally for academic research and experimentation only. I will comply with the SAM 3 license, Meta policies, and Hugging Face terms of use. I will not redistribute the model weights or provide unauthorized access to them.
I would sincerely appreciate it if you could manually review my application or reset my rejected access status so that I can submit a corrected application.
If my previous application was rejected due to incomplete or unclear information, I would be happy to provide additional details.
My information:
Hugging Face username: Windiff
Name: Lu Xiangyun
Affiliation: Peking University
Research area: Robotics and Embodied AI
Thank you very much for your time and consideration.
Best regards,
Lu Xiangyun