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 for Model Access / Academic Research on Few-shot & Reasoning Segmentation (BIT)
Dear SAM 3 Authors and Maintainers,
I am a master's graduate student from the School of Computer Science and Technology at Beijing Institute of Technology (BIT), working on computer vision and visual foundation models.
I would like to respectfully request access to the SAM 3 model checkpoints (facebook/sam3).
Research Purpose:
My current research focuses on exploring promptable segmentation under data-scarce and complex reasoning scenarios. Specifically, I plan to:
- Conduct research on Few-shot Semantic Segmentation and Reasoning Segmentation based on SAM 3's architecture.
- Evaluate and adapt SAM 3's open-vocabulary concept segmentation capabilities for complex prompt understanding and few-shot adaptation tasks.
Compliance & Usage:
- The model checkpoints will be used strictly for non-commercial, academic research purposes.
- I will adhere fully to Meta's SAM 3 License and research usage terms.
My previous request was unfortunately rejected, possibly due to incomplete profile information. I would be extremely grateful if you could kindly reconsider and grant access to my account.
Thank you very much for your time, consideration, and contribution to the open-source community!
Best regards,
Yukun An
School of Computer Science and Technology
Beijing Institute of Technology (BIT)
Email: 312024049954@bit.edu.cn