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
SAM 3 Access Request — Coastal Image Segmentation for Nautiq (pepe334)
Subject: SAM 3 access request — coastal imagery segmentation for Nautiq
Hello Meta / SAM 3 maintainers,
My Hugging Face username is pepe334. I would appreciate your review of my access request to facebook/sam3.
I am developing Nautiq, an early-stage application exploring recognizable, navigable representations of coastal locations. Our immediate goal is to evaluate SAM 3’s segmentation capabilities on aerial imagery of breakwaters, starting with a small test area in Dénia, Spain.
The proposed evaluation would focus on:
- Separating visible concrete blocks, water, and coastal structures.
- Comparing segmentation results with our existing MobileSAM baseline.
- Assessing segmentation quality, failure cases, and inference resource requirements.
This initial evaluation would use pretrained inference on a small set of imagery we are authorized to process. Nautiq may become a commercial product; this request is for technical evaluation before any production integration.
We will comply with the applicable SAM 3 license and usage conditions and will not redistribute model weights without authorization.
Please let me know if further information is required to assess this request.
Thank you for your time and consideration.
Best regards,
pepe334
Nautiq project