Model weights · SAM

Download SAM weights

The original promptable segmentation model, trained on a billion masks, that outlines any object in an image from a simple point or box prompt. Pick a size below and download the official .pt checkpoint in one click, with published parameters, size and license all in view.

Apache-2.0AGPL-3.0 via UltralyticsMeta AI · 2023 · Ultralytics (PyTorch)

SAM checkpoints, pick a size and download

The original promptable segmentation model, trained on a billion masks, that outlines any object in an image from a simple point or box prompt.

Using it with the Ultralytics package?

The SAM weights are Apache-2.0, but the Ultralytics package that most people load them with is AGPL-3.0. Running SAM through Ultralytics puts your own project under AGPL-3.0 too, which means publishing your source if you distribute it or offer it over a network. An Ultralytics Enterprise license removes that. The Apache-2.0 terms only cover you if you run the weights outside the Ultralytics package.

  • SAM base
    sam_b.pt
    Params
    93.7M
    Size
    358 MB
  • SAM large
    sam_l.pt
    Params
    312.3M
    Size
    1250 MB

Figures are published by the authors. Weights host: github.com. Clicking Download verifies the file and starts it straight from the official CDN.

How to load SAM weights

Install the Ultralytics package, then point the loader at the checkpoint. It downloads automatically on first use, or you can pass the local path to the file you downloaded above.

pip install ultralytics
from ultralytics import SAM

# Downloads on first use, or pass a local path to your .pt file
model = SAM("sam_b.pt")
results = model("image.jpg")

Want to see the architecture? Export to ONNX and open it in the ONNX visualizer for a labelled diagram with shapes and parameter counts.

Open visualizer →

Choosing a SAM checkpoint

Read off the table above: what each step up the size ladder costs and returns for this family specifically, where it stops being worth it, and what the license actually permits.

What the size ladder buys

SAM ships 2 checkpoints, from SAM base at 93.7M parameters to SAM large at 312.3M, 3.3× the model. These are promptable models with no single comparable accuracy figure, so the table lists parameters and download size rather than inventing one. Size and throughput are the axis you choose on: pick the smallest checkpoint that holds up on your own data rather than the largest you can fit.

Licensing, in practice

The weights are Apache-2.0. More importantly, the usual way to run SAM is through the `ultralytics` package, which is AGPL-3.0, and that reaches your application even though the weights themselves are not. A permissive badge on the weights is not permission to ship closed-source over that path. Apache-2.0 on the weights is commercial-friendly and needs only attribution.

About SAM

The original Segment Anything Model (Meta AI) segments any object in an image from a point, box or mask prompt, trained on the 1-billion-mask SA-1B dataset for strong zero-shot generalization. It is the foundation the whole SAM ecosystem builds on; for video and speed, prefer SAM 2.1 or the distilled MobileSAM / FastSAM below.

Author
Meta AI
Released
2023
Tasks
Segment
Framework
Ultralytics (PyTorch)
Input size
1024px
License
Apache-2.0
Resources
DocsGitHubPaperLicense

More model weights to download

Each page lists every checkpoint with accuracy, parameters, license and a one-click download.

See all 19 model families →

YOLO tracker FPS benchmarks →Get in touch →
Blog

From the blog

Tutorials, code, and notes on computer vision, deep learning, and applied AI.

View all posts →

Frequently asked questions

How do I download SAM weights?
Click the Download button next to any variant in the table above; we verify the file and start it straight from the official CDN. You can also let the loader fetch it automatically on first use with `model = SAM("sam_b.pt")`. All 2 SAM checkpoints are hosted officially.
Is SAM free for commercial use?
SAM is released under Apache-2.0. Apache-2.0 is permissive and commercial-friendly: you can use it in closed-source products with attribution. Always confirm against the linked license text.
Which SAM model size should I use?
Start with SAM base, the smallest and fastest, ideal for prototyping, edge and CPU. Move up the ladder only when you need more accuracy and have the compute for it.