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
Intended use / access request
I am requesting access to facebook/sam3 and facebook/dinov3-vitl16-pretrain-lvd1689m for non-commercial research and engineering evaluation.
Project: SimFoundry (NVlabs open-source Real2Sim / scene reconstruction pipeline)
Use case:
SAM 3: instance / ground-plane segmentation and object decomposition stages in a video-to-simulation reconstruction pipeline.
DINOv3: optional visual feature backbone for mesh-generation backends (e.g. Pixal3D-related paths), when enabled.
Weights will be downloaded via the official Hugging Face Hub with an authenticated token, used locally for inference in our research/evaluation environment only. I will comply with Meta’s model license and gated-access terms, and will not redistribute the weights, host public mirrors, or use the models for prohibited purposes.
Contact email: gcc@vip.qq.com
HF username: virtual2026
Please approve access so we can run the officially documented SimFoundry installation and reconstruction workflow.