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 manual review of rejected SAM3 access
Hello SAM 3 team,
My Hugging Face username is FFFqm. My previous request to access facebook/sam3 was rejected, and I would like to respectfully request a manual review or a reset of my access request.
I am currently conducting academic research in embodied AI, with a particular interest in agentic embodied intelligence systems. I am working on reproducing OpenETA, an agentic embodied AI framework in which SAM 3 is used as one of the visual perception tools.
As part of this reproduction work, I plan to use SAM 3 for object segmentation and visual grounding, so that the agent can identify and localize task-relevant objects in visual observations and use this information during downstream planning and interaction.
My goal is to reproduce and evaluate the OpenETA framework and then investigate possible improvements to its perception and agentic decision-making pipeline. The model will only be used for non-commercial academic research, using public or non-sensitive research data.
I will comply with the applicable SAM 3 license terms and usage requirements. Could you please reconsider my request and either reset my access request or grant access to my Hugging Face account?
Please let me know if any additional information is required.
Thank you very much for your time and consideration.