Skip to content
imaged

Segment everything

FastSAM-x

The heavy FastSAM backbone for higher recall on small or cluttered objects.

Apache-2.0
Explore model

About this model

FastSAM-x is the extra-large variant of CASIA-IVA-Lab's FastSAM. Same I/O contract as fastsam-s, swapping the YOLOv8s backbone for YOLOv8x. The build is around 289 MB on disk.

Use it for batch and offline jobs where you can spend latency for higher recall on small instances. Stick with fastsam-s for live feed.

What you can build

  • Recover small objects FastSAM-s misses in cluttered scenes.
  • Mask every object on a product page for a marketplace.
  • Run an offline annotation pipeline where recall matters more than speed.
  • Generate dense instance masks for training a downstream model.

Variants & downloads

Choose the build for your app

Model files are being prepared.

Explore the model’s examples, source links, and integration notes while download variants are added.

Example inference

Input
A hatchback photographed on a city street.
Output
The same car with denser instance masks than the small variant.

Performance

Results depend on the device, runtime, and test conditions. Each measurement keeps its own context.

M1 Max, macOS 15, ONNX Runtime 1.17

See original catalog hardware notes · CPU

3.87 s

Warm inference

Catalog reference · variant unspecifiedCold start (ms): 3904.45

Reference result from the existing catalog; conversion variant and measurement date were not recorded.

M1 Max, macOS 15, ONNX Runtime 1.17

See original catalog hardware notes · CoreML

104.8 ms

Warm inference

Catalog reference · variant unspecifiedCold start (ms): 187.05

Reference result from the existing catalog; conversion variant and measurement date were not recorded.

RTX 2060, Ubuntu 24.04, ONNX Runtime 1.24.3

See original catalog hardware notes · CUDA

119.6 ms

Warm inference

Catalog reference · variant unspecified

Reference result from the existing catalog; conversion variant and measurement date were not recorded.

Integration examples

Vision SDK integration

cpp

#include "imaged/sdk.hpp"
#include "imaged/families/fastsam/fastsam_family.hpp"

imaged::SDK::AI ai;
auto& config = ai.getOptions();
config.set("fastsam.model", "weights/fastsam_x.onnx");

imaged::SDK::FastsamFamily family;
family.load(config);

imaged::SDK::Image image;
image.load("scene.jpg");

auto result = std::get<imaged::SDK::DetectionResult>(family.run({&image}));