← All notes

Summary

Segment Anything (2023) released a foundation model and dataset for promptable image segmentation. You can point, box, or otherwise prompt the model to segment an object, and that interface quickly became a piece of vision infrastructure.

What it introduced: the Segment Anything Model (SAM), a large promptable segmentation model, and the SA-1B dataset used to train it, with a design aimed at zero-shot transfer to new image distributions.

What it led to: widespread use of SAM as a component in labeling tools, vision agents, and later segmentation models. Live notes: ao-evals-multimodal-agents.

Use it when you need a promptable segmentation component rather than a closed, dataset-specific mask model. The dataset release is part of why it became infrastructure so quickly. Where to read it: https://arxiv.org/abs/2304.02643 (v1, 5 Apr 2023).

The paper is posted on arXiv under the arXiv non-exclusive distribution license. This note links to it rather than reproducing it. The note itself is CC BY 4.0.

License: this note is CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Pointer by iinventedlight (AI-assisted). Paper license: arXiv non-exclusive distribution; this note links to it and does not reproduce it.

@misc{iil_note_xxm9fgg0,
  author = {{'iinventedlight (AI-assisted)'}},
  title = {{Segment Anything}},
  howpublished = {iinventedlight Library},
  year = {2026},
  date = {2026-11-11},
  url = {https://iinventedlight.com/library/notes/2304-02643-segment-anything/},
  note = {Permanent ID iil:note:xxm9fgg0. License: CC-BY-4.0.},
}

Markdown

Spotted a mistake? Use the MCP feedback tool or email hello@iinventedlight.com.

For language models and agents: start at /llms.txt | /agents.md | /.well-known/agent-library.json. This page in Markdown: /library/notes/2304-02643-segment-anything.md.