Prediction

ID

69yq01d3bnrgm0ch16b8dv2bvw

Status

Succeeded

Source

Web

Hardware

A40 (Large)

Total duration

16.3s

Created

12 months ago

Input

image
mask_limit: 2
mask_2_mask
crop_n_layers: 1
box_nms_thresh: 0.7
points_per_side: 64
pred_iou_thresh: 0.7
multimask_output
points_per_batch: 128
min_mask_region_area: 25
stability_score_offset: 0.7
stability_score_thresh: 0.92
crop_n_points_downscale_factor: 2

{
  "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
  "mask_limit": 2,
  "mask_2_mask": true,
  "crop_n_layers": 1,
  "box_nms_thresh": 0.7,
  "points_per_side": 64,
  "pred_iou_thresh": 0.7,
  "multimask_output": false,
  "points_per_batch": 128,
  "min_mask_region_area": 25,
  "stability_score_offset": 0.7,
  "stability_score_thresh": 0.92,
  "crop_n_points_downscale_factor": 2
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";
import fs from "node:fs";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
  {
    input: {
      image: "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
      mask_limit: 2,
      mask_2_mask: true,
      crop_n_layers: 1,
      box_nms_thresh: 0.7,
      points_per_side: 64,
      pred_iou_thresh: 0.7,
      multimask_output: false,
      points_per_batch: 128,
      min_mask_region_area: 25,
      stability_score_offset: 0.7,
      stability_score_thresh: 0.92,
      crop_n_points_downscale_factor: 2
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
    input={
        "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
        "mask_limit": 2,
        "mask_2_mask": True,
        "crop_n_layers": 1,
        "box_nms_thresh": 0.7,
        "points_per_side": 64,
        "pred_iou_thresh": 0.7,
        "multimask_output": False,
        "points_per_batch": 128,
        "min_mask_region_area": 25,
        "stability_score_offset": 0.7,
        "stability_score_thresh": 0.92,
        "crop_n_points_downscale_factor": 2
    }
)

# To access the file URL:
print(output[0].url())
#=> "http://example.com"

# To write the file to disk:
with open("my-image.png", "wb") as file:
    file.write(output[0].read())

To learn more, take a look at the guide on getting started with Python.

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
    "input": {
      "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
      "mask_limit": 2,
      "mask_2_mask": true,
      "crop_n_layers": 1,
      "box_nms_thresh": 0.7,
      "points_per_side": 64,
      "pred_iou_thresh": 0.7,
      "multimask_output": false,
      "points_per_batch": 128,
      "min_mask_region_area": 25,
      "stability_score_offset": 0.7,
      "stability_score_thresh": 0.92,
      "crop_n_points_downscale_factor": 2
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/lucataco/segment-anything-2@sha256:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0 \
  -i 'image="https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png"' \
  -i 'mask_limit=2' \
  -i 'mask_2_mask=true' \
  -i 'crop_n_layers=1' \
  -i 'box_nms_thresh=0.7' \
  -i 'points_per_side=64' \
  -i 'pred_iou_thresh=0.7' \
  -i 'multimask_output=false' \
  -i 'points_per_batch=128' \
  -i 'min_mask_region_area=25' \
  -i 'stability_score_offset=0.7' \
  -i 'stability_score_thresh=0.92' \
  -i 'crop_n_points_downscale_factor=2'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/segment-anything-2@sha256:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
      "mask_limit": 2,
      "mask_2_mask": true,
      "crop_n_layers": 1,
      "box_nms_thresh": 0.7,
      "points_per_side": 64,
      "pred_iou_thresh": 0.7,
      "multimask_output": false,
      "points_per_batch": 128,
      "min_mask_region_area": 25,
      "stability_score_offset": 0.7,
      "stability_score_thresh": 0.92,
      "crop_n_points_downscale_factor": 2
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Generated in

16.2 seconds

lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0

Prediction

Model

ID

wbc93f7z8xrgp0ch16barfhcsg

Status

Succeeded

Source

Web

Hardware

A40 (Large)

Total duration

19.5s

Created

12 months ago

Input

image
mask_limit: 3
mask_2_mask
crop_n_layers: 1
box_nms_thresh: 0.7
points_per_side: 64
pred_iou_thresh: 0.7
multimask_output
points_per_batch: 128
min_mask_region_area: 25
stability_score_offset: 0.7
stability_score_thresh: 0.92
crop_n_points_downscale_factor: 2

{
  "image": "https://replicate.delivery/pbxt/LMUHtIfEyU0uyohPBOccyITUuA7N4UJaS9Qe3zY4sMfrclBT/newt.png",
  "mask_limit": 3,
  "mask_2_mask": true,
  "crop_n_layers": 1,
  "box_nms_thresh": 0.7,
  "points_per_side": 64,
  "pred_iou_thresh": 0.7,
  "multimask_output": false,
  "points_per_batch": 128,
  "min_mask_region_area": 25,
  "stability_score_offset": 0.7,
  "stability_score_thresh": 0.92,
  "crop_n_points_downscale_factor": 2
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";
import fs from "node:fs";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
  {
    input: {
      image: "https://replicate.delivery/pbxt/LMUHtIfEyU0uyohPBOccyITUuA7N4UJaS9Qe3zY4sMfrclBT/newt.png",
      mask_limit: 3,
      mask_2_mask: true,
      crop_n_layers: 1,
      box_nms_thresh: 0.7,
      points_per_side: 64,
      pred_iou_thresh: 0.7,
      multimask_output: false,
      points_per_batch: 128,
      min_mask_region_area: 25,
      stability_score_offset: 0.7,
      stability_score_thresh: 0.92,
      crop_n_points_downscale_factor: 2
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
    input={
        "image": "https://replicate.delivery/pbxt/LMUHtIfEyU0uyohPBOccyITUuA7N4UJaS9Qe3zY4sMfrclBT/newt.png",
        "mask_limit": 3,
        "mask_2_mask": True,
        "crop_n_layers": 1,
        "box_nms_thresh": 0.7,
        "points_per_side": 64,
        "pred_iou_thresh": 0.7,
        "multimask_output": False,
        "points_per_batch": 128,
        "min_mask_region_area": 25,
        "stability_score_offset": 0.7,
        "stability_score_thresh": 0.92,
        "crop_n_points_downscale_factor": 2
    }
)

# To access the file URL:
print(output[0].url())
#=> "http://example.com"

# To write the file to disk:
with open("my-image.png", "wb") as file:
    file.write(output[0].read())

To learn more, take a look at the guide on getting started with Python.

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
    "input": {
      "image": "https://replicate.delivery/pbxt/LMUHtIfEyU0uyohPBOccyITUuA7N4UJaS9Qe3zY4sMfrclBT/newt.png",
      "mask_limit": 3,
      "mask_2_mask": true,
      "crop_n_layers": 1,
      "box_nms_thresh": 0.7,
      "points_per_side": 64,
      "pred_iou_thresh": 0.7,
      "multimask_output": false,
      "points_per_batch": 128,
      "min_mask_region_area": 25,
      "stability_score_offset": 0.7,
      "stability_score_thresh": 0.92,
      "crop_n_points_downscale_factor": 2
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/lucataco/segment-anything-2@sha256:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0 \
  -i 'image="https://replicate.delivery/pbxt/LMUHtIfEyU0uyohPBOccyITUuA7N4UJaS9Qe3zY4sMfrclBT/newt.png"' \
  -i 'mask_limit=3' \
  -i 'mask_2_mask=true' \
  -i 'crop_n_layers=1' \
  -i 'box_nms_thresh=0.7' \
  -i 'points_per_side=64' \
  -i 'pred_iou_thresh=0.7' \
  -i 'multimask_output=false' \
  -i 'points_per_batch=128' \
  -i 'min_mask_region_area=25' \
  -i 'stability_score_offset=0.7' \
  -i 'stability_score_thresh=0.92' \
  -i 'crop_n_points_downscale_factor=2'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/segment-anything-2@sha256:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/LMUHtIfEyU0uyohPBOccyITUuA7N4UJaS9Qe3zY4sMfrclBT/newt.png",
      "mask_limit": 3,
      "mask_2_mask": true,
      "crop_n_layers": 1,
      "box_nms_thresh": 0.7,
      "points_per_side": 64,
      "pred_iou_thresh": 0.7,
      "multimask_output": false,
      "points_per_batch": 128,
      "min_mask_region_area": 25,
      "stability_score_offset": 0.7,
      "stability_score_thresh": 0.92,
      "crop_n_points_downscale_factor": 2
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Generated in

19.4 seconds

lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0

Prediction

Model

ID

qrm2b1mzg5rgp0ch16bs5kvg38

Status

Succeeded

Source

Web

Hardware

A40 (Large)

Total duration

18.1s

Created

12 months ago

Input

image
mask_limit: 2
mask_2_mask
crop_n_layers: 1
box_nms_thresh: 0.7
points_per_side: 64
pred_iou_thresh: 0.7
multimask_output
points_per_batch: 128
min_mask_region_area: 25
stability_score_offset: 0.7
stability_score_thresh: 0.92
crop_n_points_downscale_factor: 2

{
  "image": "https://replicate.delivery/pbxt/LMUBwfp4nT123o6Aw5S3fpcQdUD4W6YQtAuVLfjXmvXaJeBP/cars.png",
  "mask_limit": 2,
  "mask_2_mask": true,
  "crop_n_layers": 1,
  "box_nms_thresh": 0.7,
  "points_per_side": 64,
  "pred_iou_thresh": 0.7,
  "multimask_output": false,
  "points_per_batch": 128,
  "min_mask_region_area": 25,
  "stability_score_offset": 0.7,
  "stability_score_thresh": 0.92,
  "crop_n_points_downscale_factor": 2
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";
import fs from "node:fs";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
  {
    input: {
      image: "https://replicate.delivery/pbxt/LMUBwfp4nT123o6Aw5S3fpcQdUD4W6YQtAuVLfjXmvXaJeBP/cars.png",
      mask_limit: 2,
      mask_2_mask: true,
      crop_n_layers: 1,
      box_nms_thresh: 0.7,
      points_per_side: 64,
      pred_iou_thresh: 0.7,
      multimask_output: false,
      points_per_batch: 128,
      min_mask_region_area: 25,
      stability_score_offset: 0.7,
      stability_score_thresh: 0.92,
      crop_n_points_downscale_factor: 2
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
    input={
        "image": "https://replicate.delivery/pbxt/LMUBwfp4nT123o6Aw5S3fpcQdUD4W6YQtAuVLfjXmvXaJeBP/cars.png",
        "mask_limit": 2,
        "mask_2_mask": True,
        "crop_n_layers": 1,
        "box_nms_thresh": 0.7,
        "points_per_side": 64,
        "pred_iou_thresh": 0.7,
        "multimask_output": False,
        "points_per_batch": 128,
        "min_mask_region_area": 25,
        "stability_score_offset": 0.7,
        "stability_score_thresh": 0.92,
        "crop_n_points_downscale_factor": 2
    }
)

# To access the file URL:
print(output[0].url())
#=> "http://example.com"

# To write the file to disk:
with open("my-image.png", "wb") as file:
    file.write(output[0].read())

To learn more, take a look at the guide on getting started with Python.

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
    "input": {
      "image": "https://replicate.delivery/pbxt/LMUBwfp4nT123o6Aw5S3fpcQdUD4W6YQtAuVLfjXmvXaJeBP/cars.png",
      "mask_limit": 2,
      "mask_2_mask": true,
      "crop_n_layers": 1,
      "box_nms_thresh": 0.7,
      "points_per_side": 64,
      "pred_iou_thresh": 0.7,
      "multimask_output": false,
      "points_per_batch": 128,
      "min_mask_region_area": 25,
      "stability_score_offset": 0.7,
      "stability_score_thresh": 0.92,
      "crop_n_points_downscale_factor": 2
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/lucataco/segment-anything-2@sha256:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0 \
  -i 'image="https://replicate.delivery/pbxt/LMUBwfp4nT123o6Aw5S3fpcQdUD4W6YQtAuVLfjXmvXaJeBP/cars.png"' \
  -i 'mask_limit=2' \
  -i 'mask_2_mask=true' \
  -i 'crop_n_layers=1' \
  -i 'box_nms_thresh=0.7' \
  -i 'points_per_side=64' \
  -i 'pred_iou_thresh=0.7' \
  -i 'multimask_output=false' \
  -i 'points_per_batch=128' \
  -i 'min_mask_region_area=25' \
  -i 'stability_score_offset=0.7' \
  -i 'stability_score_thresh=0.92' \
  -i 'crop_n_points_downscale_factor=2'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/segment-anything-2@sha256:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/LMUBwfp4nT123o6Aw5S3fpcQdUD4W6YQtAuVLfjXmvXaJeBP/cars.png",
      "mask_limit": 2,
      "mask_2_mask": true,
      "crop_n_layers": 1,
      "box_nms_thresh": 0.7,
      "points_per_side": 64,
      "pred_iou_thresh": 0.7,
      "multimask_output": false,
      "points_per_batch": 128,
      "min_mask_region_area": 25,
      "stability_score_offset": 0.7,
      "stability_score_thresh": 0.92,
      "crop_n_points_downscale_factor": 2
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Generated in

18.0 seconds

lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0

Prediction

Model

ID

tfxh1xt85drgm0ch16cb6pnpmw

Status

Succeeded

Source

Web

Hardware

A40 (Large)

Total duration

34.8s

Created

12 months ago

Input

image
mask_limit: -1
mask_2_mask
crop_n_layers: 1
box_nms_thresh: 0.7
points_per_side: 64
pred_iou_thresh: 0.7
multimask_output
points_per_batch: 128
min_mask_region_area: 25
stability_score_offset: 0.7
stability_score_thresh: 0.92
crop_n_points_downscale_factor: 2

{
  "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
  "mask_limit": -1,
  "mask_2_mask": true,
  "crop_n_layers": 1,
  "box_nms_thresh": 0.7,
  "points_per_side": 64,
  "pred_iou_thresh": 0.7,
  "multimask_output": false,
  "points_per_batch": 128,
  "min_mask_region_area": 25,
  "stability_score_offset": 0.7,
  "stability_score_thresh": 0.92,
  "crop_n_points_downscale_factor": 2
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";
import fs from "node:fs";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
  {
    input: {
      image: "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
      mask_limit: -1,
      mask_2_mask: true,
      crop_n_layers: 1,
      box_nms_thresh: 0.7,
      points_per_side: 64,
      pred_iou_thresh: 0.7,
      multimask_output: false,
      points_per_batch: 128,
      min_mask_region_area: 25,
      stability_score_offset: 0.7,
      stability_score_thresh: 0.92,
      crop_n_points_downscale_factor: 2
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
    input={
        "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
        "mask_limit": -1,
        "mask_2_mask": True,
        "crop_n_layers": 1,
        "box_nms_thresh": 0.7,
        "points_per_side": 64,
        "pred_iou_thresh": 0.7,
        "multimask_output": False,
        "points_per_batch": 128,
        "min_mask_region_area": 25,
        "stability_score_offset": 0.7,
        "stability_score_thresh": 0.92,
        "crop_n_points_downscale_factor": 2
    }
)

# To access the file URL:
print(output[0].url())
#=> "http://example.com"

# To write the file to disk:
with open("my-image.png", "wb") as file:
    file.write(output[0].read())

To learn more, take a look at the guide on getting started with Python.

Run lucataco/segment-anything-2 using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "lucataco/segment-anything-2:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0",
    "input": {
      "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
      "mask_limit": -1,
      "mask_2_mask": true,
      "crop_n_layers": 1,
      "box_nms_thresh": 0.7,
      "points_per_side": 64,
      "pred_iou_thresh": 0.7,
      "multimask_output": false,
      "points_per_batch": 128,
      "min_mask_region_area": 25,
      "stability_score_offset": 0.7,
      "stability_score_thresh": 0.92,
      "crop_n_points_downscale_factor": 2
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/lucataco/segment-anything-2@sha256:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0 \
  -i 'image="https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png"' \
  -i 'mask_limit=-1' \
  -i 'mask_2_mask=true' \
  -i 'crop_n_layers=1' \
  -i 'box_nms_thresh=0.7' \
  -i 'points_per_side=64' \
  -i 'pred_iou_thresh=0.7' \
  -i 'multimask_output=false' \
  -i 'points_per_batch=128' \
  -i 'min_mask_region_area=25' \
  -i 'stability_score_offset=0.7' \
  -i 'stability_score_thresh=0.92' \
  -i 'crop_n_points_downscale_factor=2'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/segment-anything-2@sha256:be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
      "mask_limit": -1,
      "mask_2_mask": true,
      "crop_n_layers": 1,
      "box_nms_thresh": 0.7,
      "points_per_side": 64,
      "pred_iou_thresh": 0.7,
      "multimask_output": false,
      "points_per_batch": 128,
      "min_mask_region_area": 25,
      "stability_score_offset": 0.7,
      "stability_score_thresh": 0.92,
      "crop_n_points_downscale_factor": 2
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

{
  "completed_at": "2024-07-31T16:21:35.415110Z",
  "created_at": "2024-07-31T16:21:00.587000Z",
  "data_removed": false,
  "error": null,
  "id": "tfxh1xt85drgm0ch16cb6pnpmw",
  "input": {
    "image": "https://replicate.delivery/pbxt/LMUHQoCnNzN15MpwWLPJUVF96g3zxer6p3dNTHtVrNxqMMe0/socer.png",
    "mask_limit": -1,
    "mask_2_mask": true,
    "crop_n_layers": 1,
    "box_nms_thresh": 0.7,
    "points_per_side": 64,
    "pred_iou_thresh": 0.7,
    "multimask_output": false,
    "points_per_batch": 128,
    "min_mask_region_area": 25,
    "stability_score_offset": 0.7,
    "stability_score_thresh": 0.92,
    "crop_n_points_downscale_factor": 2
  },
  "logs": null,
  "metrics": {
    "predict_time": 34.65755729,
    "total_time": 34.82811
  },
  "output": [
    "https://replicate.delivery/pbxt/wMe3b6Ajiq1LICMqezYf6Q5x9XykyEKLisda25V1pRF4zCcmA/mask_0.png",
    "https://replicate.delivery/pbxt/CXst4oQ095rKCdfgZcKRPrEtzzgFP0ESyIcf5ZqHjOB8ZBOTA/mask_1.png",
    "https://replicate.delivery/pbxt/9jBbLFZtDezSH65Z2vF91Zsl9QFNj3MsrWDMXrfslVY9ZBOTA/mask_2.png",
    "https://replicate.delivery/pbxt/O2LCd6MmTLoaKhxj2yS8JlY7eck6S8RMZm5hQL9WmiseZBOTA/mask_3.png",
    "https://replicate.delivery/pbxt/obfzpkOfBjjd30RwRwwuaaK2qd3Oi9jvZGffwW5pzj00nF4MB/mask_4.png",
    "https://replicate.delivery/pbxt/1kehgrKHnHQkMSD1iK8zTeaEUWIEGtmPUpKx4bP8hAz9ZBOTA/mask_5.png",
    "https://replicate.delivery/pbxt/Tcl8ELTZR4LQCxNxcBCJ2XvjAVtw08qCeCqIhOLxzqKfZBOTA/mask_6.png",
    "https://replicate.delivery/pbxt/BEfcJtd4rm1e7EZrh4CntmJ4HrcSeU9HXQ9OLXOOB4l8zCcmA/mask_7.png",
    "https://replicate.delivery/pbxt/YskqOVH8TaKXL5Gp9mFrvIH09lR1MleE5t9qONNzJ3fezCcmA/mask_8.png",
    "https://replicate.delivery/pbxt/0eNrv8Ov3Py6Qa5Esn8ff1ysXE59CXMzka6X6FvfPYq7nF4MB/mask_9.png",
    "https://replicate.delivery/pbxt/ees0yjuow0tgjkLmVjkDKViNBl5PR3Rlw8LPsefzDJZ9nF4MB/mask_10.png",
    "https://replicate.delivery/pbxt/jST7uGKsY1aXGhGOW8mNeZs7CPrcrw2UKbW1BGIkk2mfZBOTA/mask_11.png",
    "https://replicate.delivery/pbxt/VsSUbxxWUnaDFxuf8bsm7EfCdhIYeob4mYs4RGkSexXfPLwZC/mask_12.png",
    "https://replicate.delivery/pbxt/QZ4IDjw0DT6jHtRn9Nk8OuVHmkfD1nho9hTXJbM0OqLAtAnJA/mask_13.png",
    "https://replicate.delivery/pbxt/h6fXVBU7OcW4LSuBnwf0iaf3XUxV1X4gAOW1cF06AhvA0CcmA/mask_14.png",
    "https://replicate.delivery/pbxt/74WV8NsmwY4hA5YTzJFVS39TD8sX0lxBtSTfzeBmm4vAaBOTA/mask_15.png",
    "https://replicate.delivery/pbxt/I3NuM6LEUoo9AFhPrfC9sNtdIv6S8AbkmIL6CFIofu1AaBOTA/mask_16.png",
    "https://replicate.delivery/pbxt/edA27QurYKzpbamC1eGejgdFfeYM3pCH8m25fiTDaYVagWgzE/mask_17.png",
    "https://replicate.delivery/pbxt/jyiK5oJVC7biKNKDv2vihQ3x7xf2CV0sZ2t0rGMef4RC0CcmA/mask_18.png",
    "https://replicate.delivery/pbxt/LBJfv1EZf6vVKEFGOz5sOufmVfPs8ifzNPiPmIhQJLQMQLwZC/mask_19.png",
    "https://replicate.delivery/pbxt/B3V50yi06eUyDCY7Ci8BfyYxkGsXFSkYcEpDLXWGshYBaBOTA/mask_20.png",
    "https://replicate.delivery/pbxt/SSGhfqoeXWiD7EJDte8kr43PhTB8JkuCdBqzypqEOfzIoF4MB/mask_21.png",
    "https://replicate.delivery/pbxt/LzLAqvH1MVJyLJWDmq28er3ocy7Sa7Zi20OYXSdI6jTBtAnJA/mask_22.png",
    "https://replicate.delivery/pbxt/jcfy8muZYFS4OKf9E6ycP9qZfSLKeFOFOG3kdl16gJNIoF4MB/mask_23.png",
    "https://replicate.delivery/pbxt/jyRKKcnYsPaoI9hldtJAPri6wKZfXaOT5unfbOeWxz2F0CcmA/mask_24.png",
    "https://replicate.delivery/pbxt/tRgwsUlAtJLXEJw2WYBkScH2ezyWsNs4cBku1e1vdmSDaBOTA/mask_25.png",
    "https://replicate.delivery/pbxt/B6eylHnFmvSXZKPIu1c1DfUAevgiJMdwFJULhopFmUNH0CcmA/mask_26.png",
    "https://replicate.delivery/pbxt/3cZMdOfGWv22DCsruffA3d8IftBILWqO6vz8DP32spRMoF4MB/mask_27.png",
    "https://replicate.delivery/pbxt/PmHRItF8AmqWJdhpvhiVRNYSydy2nmPFpyq5oPCo57zgWgzE/mask_28.png",
    "https://replicate.delivery/pbxt/ZKmjwepz2MTCW6XY3jNzvDGnMkl1UfBD1Sp8mUkxT68EaBOTA/mask_29.png",
    "https://replicate.delivery/pbxt/a2Z77s68sfU3AqQsHIwtkkDF9NqXjK4yO9qIf67XsWvEaBOTA/mask_30.png",
    "https://replicate.delivery/pbxt/5RaeY2joANQHI6Qf3t8cSui9GJrVmoBWbfZY8eoBBq4ToF4MB/mask_31.png",
    "https://replicate.delivery/pbxt/wse9gcXnsw1CUiq095d9KgLpwgfisgwCV45pHUodEuHEaBOTA/mask_32.png",
    "https://replicate.delivery/pbxt/h1aSpDJl0qKuCJ6VGUnexrjyaLRP5NzjSRtzNPULM7iCtAnJA/mask_33.png",
    "https://replicate.delivery/pbxt/X4JfR52yafvwOU65Yewb8liKf99ch3qFZmGVOTrine3sQLwZC/mask_34.png",
    "https://replicate.delivery/pbxt/MaUmdYiN6l71OdbmfOJqsdQBm7pp0h2b6a9D4TLf2bYFaBOTA/mask_35.png",
    "https://replicate.delivery/pbxt/Bfr3SpXOQF3wRae1kqDDxhksKdNgASdzHkrrl2mZqdzGaBOTA/mask_36.png",
    "https://replicate.delivery/pbxt/cXz8iEnmFV7LNRhWzGy6nVNTiJvw0EXAvKJnDakrEDkhWgzE/mask_37.png",
    "https://replicate.delivery/pbxt/OEumfr9VyeujlkpmSTQ9g6qm3tk6LydvBgyJjIfDSbVN0CcmA/mask_38.png",
    "https://replicate.delivery/pbxt/maDseYNhVnS8TyrYvPP88pRzEzkPU0fW9CVznTdiiBSGaBOTA/mask_39.png",
    "https://replicate.delivery/pbxt/9TEbMhLTUWbDAhaRkGZaffL5q35l3kaJKKqvof1eB4EeQLwZC/mask_40.png",
    "https://replicate.delivery/pbxt/BF6O3UwJGlJ4DRjrEO0Cneb2RhxxXh8ZVgERRt6prvoDtAnJA/mask_41.png",
    "https://replicate.delivery/pbxt/5EKiOPNMmv6AApo5e1BR4WeNwfwpXntcwRQemzgPev98QLwZC/mask_42.png",
    "https://replicate.delivery/pbxt/lRgQffpjhEhh5kampLqin8zGn6IwCy9MfkaAnAeJ0EZdoF4MB/mask_43.png",
    "https://replicate.delivery/pbxt/twnKEF3YoKqnAZ11eLmRs2qeskbTHFDLsx36WTYbaT7IaBOTA/mask_44.png",
    "https://replicate.delivery/pbxt/KBrP595FllpqLxW6dMrFf6QwVjeX00PQceIMBcC4PzIQ0CcmA/mask_45.png",
    "https://replicate.delivery/pbxt/mAsRfQPaZcyLLigx9LYpHdIBo9yK0sIjojQ670LoxQVEtAnJA/mask_46.png",
    "https://replicate.delivery/pbxt/zzYOAagFgPpBKFKH7zDK1AgyTMClcPKx2D0BNbH5wlJiWgzE/mask_47.png",
    "https://replicate.delivery/pbxt/6i3VI7b6oW4eAaGEIUbWrPdB8F9yJDKBqpufQ84ICqGJaBOTA/mask_48.png",
    "https://replicate.delivery/pbxt/5qEVJ7ZtvQbJDZcMMteJ9bhkz6V4DSfbtqy98fFmIq9S0CcmA/mask_49.png",
    "https://replicate.delivery/pbxt/jrDg99MS2CIvC5v3Ui8VR02Vg9gAhWvR4zJVpZWDgheEtAnJA/mask_50.png",
    "https://replicate.delivery/pbxt/veawjIXQjSXbNCeGVD4hwvctsNum2ZPeyni0BkzfmxqkoF4MB/mask_51.png",
    "https://replicate.delivery/pbxt/r4wtddTfPbRBayMQgX1oRR7fU9a7pM6eXEStesABI1WpoF4MB/mask_52.png",
    "https://replicate.delivery/pbxt/ghDYxm0vIspEDZ8IUqZ4XAu2pR7DWgQ9VptPCwDJi9giWgzE/mask_53.png",
    "https://replicate.delivery/pbxt/z12NWxypUh6EHh2beWULJfFgNiDxNnJGvSSqeVk38poU0CcmA/mask_54.png",
    "https://replicate.delivery/pbxt/nfUed3fSTVQ9sIvyflVvOKnOX99dVACLAoWwV3vBznWsoF4MB/mask_55.png",
    "https://replicate.delivery/pbxt/ctSrxpB0s0I1NxED624SlrHhBTXyff6JQI8i7meaeWWvoF4MB/mask_56.png",
    "https://replicate.delivery/pbxt/Y0AU8ejVfWvnzkhOMPqhBjv6HO4OPqzrrOErMgJf5NgX0CcmA/mask_57.png",
    "https://replicate.delivery/pbxt/syvObD77Iz53PpJ9Ay27SlIeKIQTW2nl2iNEoeSLDVPLaBOTA/mask_58.png",
    "https://replicate.delivery/pbxt/GBuHNzOxQ8p1IFfjnrx9ehms2lw0Xb2M56n9BY5551cMaBOTA/mask_59.png",
    "https://replicate.delivery/pbxt/yHD22wf39IWFfk7BKXGWar3Df1HnN8PSpWCwfFoPOJlxoF4MB/mask_60.png",
    "https://replicate.delivery/pbxt/LKTWdpy7MEoeYKfsTPh1IxWewf5hIQubQnh3dsWzjfCjRLwZC/mask_61.png",
    "https://replicate.delivery/pbxt/ekRjrVws3OxBNaS9MnLI6SYEafN4KLFif0HQk4niNuTY0CcmA/mask_62.png",
    "https://replicate.delivery/pbxt/BszDmwrdMfUxZCm4c4QBNBGo0Lmr0mUvMyky1SsVVdfNaBOTA/mask_63.png",
    "https://replicate.delivery/pbxt/tQCbffUfrWP7Ao7xhqgANAUKMLRIvcsSN9OMgdD8DGyb0CcmA/mask_64.png",
    "https://replicate.delivery/pbxt/hzq3AvHdPQZrIRbQhiB3fSDpm2SrJPGT9ohYhbWNtPfNaBOTA/mask_65.png",
    "https://replicate.delivery/pbxt/DrivEqNaULqSL1UidAfK6vy42pDJW5USiGZdtGttfQ8NaBOTA/mask_66.png",
    "https://replicate.delivery/pbxt/mtB8uutp6La8MBWKPW6uShQkhvu3NxaWMefvUetV5Je7oF4MB/mask_67.png",
    "https://replicate.delivery/pbxt/wpFWls3GhzZOMVOjSVU0PfNn0auyMilPWVnblntJsOHHtAnJA/mask_68.png",
    "https://replicate.delivery/pbxt/nBzcyPFK6HqLOl8SLsMe7u2PBfx41XGAhBFQrEfkhagd0CcmA/mask_69.png",
    "https://replicate.delivery/pbxt/8P6eUk903l0rLKy4icJDyVNEMhqLOhrTSjv8oYvDvVSHtAnJA/mask_70.png",
    "https://replicate.delivery/pbxt/4QaOhNXffzjAeIeUHwjgauDAb8hvCfykWzWWeFznlyL9jWgzE/mask_71.png"
  ],
  "started_at": "2024-07-31T16:21:00.757553Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/tfxh1xt85drgm0ch16cb6pnpmw",
    "cancel": "https://api.replicate.com/v1/predictions/tfxh1xt85drgm0ch16cb6pnpmw/cancel"
  },
  "version": "be7cbde9fdf0eecdc8b20ffec9dd0d1cfeace0832d4d0b58a071d993182e1be0"
}

Generated in

34.7 seconds