sepal/sdxl-inpainting – Run with an API on Replicate

sepal / sdxl-inpainting

Stable Diffusion XL specifically trained on Inpainting by huggingface

Cold

Public
9.5K runs
L40S
GitHub

Run with an API

Playground API Examples README Versions

Input

prompt

string

Shift + Return to add a new line

Input prompt

Default: "An astronaut riding a rainbow unicorn"

negative_prompt

string

Shift + Return to add a new line

Input Negative Prompt

Default: ""

image

file

Preview

Input image for img2img or inpaint mode

mask

file

Preview

Input mask for inpaint mode. Black areas will be preserved, white areas will be inpainted.

num_inference_steps

integer

(minimum: 1, maximum: 500)

Number of denoising steps

Default: 50

guidance_scale

number

(minimum: 1, maximum: 50)

Scale for classifier-free guidance

Default: 7.5

prompt_strength

number

(minimum: 0, maximum: 1)

Prompt strength when using img2img / inpaint. 1.0 corresponds to full destruction of information in image

Default: 0.8

seed

integer

Random seed. Leave blank to randomize the seed

Run this model in Node.js with one line of code:

npx create-replicate --model=sepal/sdxl-inpainting

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

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

Run sepal/sdxl-inpainting using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "sepal/sdxl-inpainting:aca001c8b137114d5e594c68f7084ae6d82f364758aab8d997b233e8ef3c4d93",
  {
    input: {
      mask: "https://replicate.delivery/pbxt/JTrIcNokOgIDb3Y5ouSvWmXf6iQAheTJG7UaTZGje2VBnbeh/overture-creations-5sI6fQgYIuo_mask.png",
      image: "https://replicate.delivery/pbxt/JTrIcA5OpankV1HZfwXyEmmZLZKparZrmGjCMIKwYLiEUyTj/overture-creations-5sI6fQgYIuo.png",
      prompt: "a tiger sitting on a park bench",
      guidance_scale: 7.5,
      negative_prompt: "",
      prompt_strength: 0.8,
      num_inference_steps: 50
    }
  }
);

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

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

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

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run sepal/sdxl-inpainting using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "sepal/sdxl-inpainting:aca001c8b137114d5e594c68f7084ae6d82f364758aab8d997b233e8ef3c4d93",
    input={
        "mask": "https://replicate.delivery/pbxt/JTrIcNokOgIDb3Y5ouSvWmXf6iQAheTJG7UaTZGje2VBnbeh/overture-creations-5sI6fQgYIuo_mask.png",
        "image": "https://replicate.delivery/pbxt/JTrIcA5OpankV1HZfwXyEmmZLZKparZrmGjCMIKwYLiEUyTj/overture-creations-5sI6fQgYIuo.png",
        "prompt": "a tiger sitting on a park bench",
        "guidance_scale": 7.5,
        "negative_prompt": "",
        "prompt_strength": 0.8,
        "num_inference_steps": 50
    }
)
print(output)

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

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run sepal/sdxl-inpainting 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": "sepal/sdxl-inpainting:aca001c8b137114d5e594c68f7084ae6d82f364758aab8d997b233e8ef3c4d93",
    "input": {
      "mask": "https://replicate.delivery/pbxt/JTrIcNokOgIDb3Y5ouSvWmXf6iQAheTJG7UaTZGje2VBnbeh/overture-creations-5sI6fQgYIuo_mask.png",
      "image": "https://replicate.delivery/pbxt/JTrIcA5OpankV1HZfwXyEmmZLZKparZrmGjCMIKwYLiEUyTj/overture-creations-5sI6fQgYIuo.png",
      "prompt": "a tiger sitting on a park bench",
      "guidance_scale": 7.5,
      "negative_prompt": "",
      "prompt_strength": 0.8,
      "num_inference_steps": 50
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2023-09-06T13:03:59.241537Z",
  "created_at": "2023-09-06T13:03:25.529771Z",
  "data_removed": false,
  "error": null,
  "id": "dbtr4mlba2g6re2fjrkx3y5w5m",
  "input": {
    "mask": "https://replicate.delivery/pbxt/JTrIcNokOgIDb3Y5ouSvWmXf6iQAheTJG7UaTZGje2VBnbeh/overture-creations-5sI6fQgYIuo_mask.png",
    "image": "https://replicate.delivery/pbxt/JTrIcA5OpankV1HZfwXyEmmZLZKparZrmGjCMIKwYLiEUyTj/overture-creations-5sI6fQgYIuo.png",
    "prompt": "a tiger sitting on a park bench",
    "guidance_scale": 7.5,
    "prompt_strength": 0.8,
    "num_inference_steps": 50
  },
  "logs": "0%|          | 0/40 [00:00<?, ?it/s]\n  2%|▎         | 1/40 [00:00<00:08,  4.79it/s]\n  5%|▌         | 2/40 [00:00<00:07,  4.89it/s]\n  8%|▊         | 3/40 [00:00<00:07,  4.92it/s]\n 10%|█         | 4/40 [00:00<00:07,  4.93it/s]\n 12%|█▎        | 5/40 [00:01<00:07,  4.93it/s]\n 15%|█▌        | 6/40 [00:01<00:06,  4.93it/s]\n 18%|█▊        | 7/40 [00:01<00:06,  4.94it/s]\n 20%|██        | 8/40 [00:01<00:06,  4.94it/s]\n 22%|██▎       | 9/40 [00:01<00:06,  4.94it/s]\n 25%|██▌       | 10/40 [00:02<00:06,  4.94it/s]\n 28%|██▊       | 11/40 [00:02<00:05,  4.94it/s]\n 30%|███       | 12/40 [00:02<00:05,  4.94it/s]\n 32%|███▎      | 13/40 [00:02<00:05,  4.94it/s]\n 35%|███▌      | 14/40 [00:02<00:05,  4.94it/s]\n 38%|███▊      | 15/40 [00:03<00:05,  4.94it/s]\n 40%|████      | 16/40 [00:03<00:04,  4.93it/s]\n 42%|████▎     | 17/40 [00:03<00:04,  4.93it/s]\n 45%|████▌     | 18/40 [00:03<00:04,  4.94it/s]\n 48%|████▊     | 19/40 [00:03<00:04,  4.95it/s]\n 50%|█████     | 20/40 [00:04<00:04,  4.96it/s]\n 52%|█████▎    | 21/40 [00:04<00:03,  4.97it/s]\n 55%|█████▌    | 22/40 [00:04<00:03,  4.96it/s]\n 57%|█████▊    | 23/40 [00:04<00:03,  4.96it/s]\n 60%|██████    | 24/40 [00:04<00:03,  4.97it/s]\n 62%|██████▎   | 25/40 [00:05<00:03,  4.97it/s]\n 65%|██████▌   | 26/40 [00:05<00:02,  4.97it/s]\n 68%|██████▊   | 27/40 [00:05<00:02,  4.97it/s]\n 70%|███████   | 28/40 [00:05<00:02,  4.96it/s]\n 72%|███████▎  | 29/40 [00:05<00:02,  4.96it/s]\n 75%|███████▌  | 30/40 [00:06<00:02,  4.96it/s]\n 78%|███████▊  | 31/40 [00:06<00:01,  4.97it/s]\n 80%|████████  | 32/40 [00:06<00:01,  4.97it/s]\n 82%|████████▎ | 33/40 [00:06<00:01,  4.96it/s]\n 85%|████████▌ | 34/40 [00:06<00:01,  4.96it/s]\n 88%|████████▊ | 35/40 [00:07<00:01,  4.97it/s]\n 90%|█████████ | 36/40 [00:07<00:00,  4.97it/s]\n 92%|█████████▎| 37/40 [00:07<00:00,  4.96it/s]\n 95%|█████████▌| 38/40 [00:07<00:00,  4.96it/s]\n 98%|█████████▊| 39/40 [00:07<00:00,  4.91it/s]\n100%|██████████| 40/40 [00:08<00:00,  4.90it/s]\n100%|██████████| 40/40 [00:08<00:00,  4.94it/s]\n/tmp/out.png",
  "metrics": {
    "predict_time": 12.744003,
    "total_time": 33.711766
  },
  "output": "https://replicate.delivery/pbxt/zK4CsmMGyn6xPl1nHWqpDfodvOtanIJUWMpDj58jBlWfqihRA/out.png",
  "started_at": "2023-09-06T13:03:46.497534Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/dbtr4mlba2g6re2fjrkx3y5w5m",
    "cancel": "https://api.replicate.com/v1/predictions/dbtr4mlba2g6re2fjrkx3y5w5m/cancel"
  },
  "version": "aca001c8b137114d5e594c68f7084ae6d82f364758aab8d997b233e8ef3c4d93"
}

Generated in

12.8 seconds

Tweak itReport

0%|          | 0/40 [00:00<?, ?it/s]
  2%|▎         | 1/40 [00:00<00:08,  4.79it/s]
  5%|▌         | 2/40 [00:00<00:07,  4.89it/s]
  8%|▊         | 3/40 [00:00<00:07,  4.92it/s]
 10%|█         | 4/40 [00:00<00:07,  4.93it/s]
 12%|█▎        | 5/40 [00:01<00:07,  4.93it/s]
 15%|█▌        | 6/40 [00:01<00:06,  4.93it/s]
 18%|█▊        | 7/40 [00:01<00:06,  4.94it/s]
 20%|██        | 8/40 [00:01<00:06,  4.94it/s]
 22%|██▎       | 9/40 [00:01<00:06,  4.94it/s]
 25%|██▌       | 10/40 [00:02<00:06,  4.94it/s]
 28%|██▊       | 11/40 [00:02<00:05,  4.94it/s]
 30%|███       | 12/40 [00:02<00:05,  4.94it/s]
 32%|███▎      | 13/40 [00:02<00:05,  4.94it/s]
 35%|███▌      | 14/40 [00:02<00:05,  4.94it/s]
 38%|███▊      | 15/40 [00:03<00:05,  4.94it/s]
 40%|████      | 16/40 [00:03<00:04,  4.93it/s]
 42%|████▎     | 17/40 [00:03<00:04,  4.93it/s]
 45%|████▌     | 18/40 [00:03<00:04,  4.94it/s]
 48%|████▊     | 19/40 [00:03<00:04,  4.95it/s]
 50%|█████     | 20/40 [00:04<00:04,  4.96it/s]
 52%|█████▎    | 21/40 [00:04<00:03,  4.97it/s]
 55%|█████▌    | 22/40 [00:04<00:03,  4.96it/s]
 57%|█████▊    | 23/40 [00:04<00:03,  4.96it/s]
 60%|██████    | 24/40 [00:04<00:03,  4.97it/s]
 62%|██████▎   | 25/40 [00:05<00:03,  4.97it/s]
 65%|██████▌   | 26/40 [00:05<00:02,  4.97it/s]
 68%|██████▊   | 27/40 [00:05<00:02,  4.97it/s]
 70%|███████   | 28/40 [00:05<00:02,  4.96it/s]
 72%|███████▎  | 29/40 [00:05<00:02,  4.96it/s]
 75%|███████▌  | 30/40 [00:06<00:02,  4.96it/s]
 78%|███████▊  | 31/40 [00:06<00:01,  4.97it/s]
 80%|████████  | 32/40 [00:06<00:01,  4.97it/s]
 82%|████████▎ | 33/40 [00:06<00:01,  4.96it/s]
 85%|████████▌ | 34/40 [00:06<00:01,  4.96it/s]
 88%|████████▊ | 35/40 [00:07<00:01,  4.97it/s]
 90%|█████████ | 36/40 [00:07<00:00,  4.97it/s]
 92%|█████████▎| 37/40 [00:07<00:00,  4.96it/s]
 95%|█████████▌| 38/40 [00:07<00:00,  4.96it/s]
 98%|█████████▊| 39/40 [00:07<00:00,  4.91it/s]
100%|██████████| 40/40 [00:08<00:00,  4.90it/s]
100%|██████████| 40/40 [00:08<00:00,  4.94it/s]
/tmp/out.png

Examples

View more examples

Run time and cost

This model costs approximately $0.0036 to run on Replicate, or 277 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 4 seconds.

Readme

This was created by Huggingface specifically for inpainting. You can read more about the model here: https://huggingface.co/diffusers/stable-diffusion-xl-1.0-inpainting-0.1