stability-ai/sdxl:610dddf0 – Run with an API on Replicate

Version

You're looking at a specific version of this model. Jump to the model overview.

stability-ai /sdxl:610dddf0

Playground API

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

Input image for img2img or inpaint mode

mask

file

Input mask for inpaint mode

width

integer

Width of output image

Default: 1024

height

integer

Height of output image

Default: 1024

num_outputs

integer

(minimum: 1, maximum: 4)

Number of images to output.

Default: 1

scheduler

string

scheduler

Default: "KarrasDPM"

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

refine

string

Which refine style to use

Default: "no_refiner"

high_noise_frac

number

(minimum: 0, maximum: 1)

for expert_ensemble_refiner, the fraction of noise to use

Default: 0.8

refine_steps

integer

for base_image_refiner, the number of steps to refine, defaults to num_inference_steps

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

npx create-replicate --model=stability-ai/sdxl

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

const output = await replicate.run(
  "stability-ai/sdxl:610dddf033f10431b1b55f24510b6009fcba23017ee551a1b9afbc4eec79e29c",
  {
    input: {
      width: 1248,
      height: 832,
      prompt: "A beautiful landscape photo",
      refine: "expert_ensemble_refiner",
      scheduler: "KarrasDPM",
      num_outputs: 1,
      guidance_scale: 7.5,
      high_noise_frac: 0.8,
      negative_prompt: "soft, blurry, ugly",
      prompt_strength: 0.8,
      num_inference_steps: 50
    }
  }
);

// 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.

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

output = replicate.run(
    "stability-ai/sdxl:610dddf033f10431b1b55f24510b6009fcba23017ee551a1b9afbc4eec79e29c",
    input={
        "width": 1248,
        "height": 832,
        "prompt": "A beautiful landscape photo",
        "refine": "expert_ensemble_refiner",
        "scheduler": "KarrasDPM",
        "num_outputs": 1,
        "guidance_scale": 7.5,
        "high_noise_frac": 0.8,
        "negative_prompt": "soft, blurry, ugly",
        "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 stability-ai/sdxl 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": "610dddf033f10431b1b55f24510b6009fcba23017ee551a1b9afbc4eec79e29c",
    "input": {
      "width": 1248,
      "height": 832,
      "prompt": "A beautiful landscape photo",
      "refine": "expert_ensemble_refiner",
      "scheduler": "KarrasDPM",
      "num_outputs": 1,
      "guidance_scale": 7.5,
      "high_noise_frac": 0.8,
      "negative_prompt": "soft, blurry, ugly",
      "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.

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/stability-ai/sdxl@sha256:610dddf033f10431b1b55f24510b6009fcba23017ee551a1b9afbc4eec79e29c \
  -i 'width=1248' \
  -i 'height=832' \
  -i 'prompt="A beautiful landscape photo"' \
  -i 'refine="expert_ensemble_refiner"' \
  -i 'scheduler="KarrasDPM"' \
  -i 'num_outputs=1' \
  -i 'guidance_scale=7.5' \
  -i 'high_noise_frac=0.8' \
  -i 'negative_prompt="soft, blurry, ugly"' \
  -i 'prompt_strength=0.8' \
  -i 'num_inference_steps=50'

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/stability-ai/sdxl@sha256:610dddf033f10431b1b55f24510b6009fcba23017ee551a1b9afbc4eec79e29c
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "width": 1248,
      "height": 832,
      "prompt": "A beautiful landscape photo",
      "refine": "expert_ensemble_refiner",
      "scheduler": "KarrasDPM",
      "num_outputs": 1,
      "guidance_scale": 7.5,
      "high_noise_frac": 0.8,
      "negative_prompt": "soft, blurry, ugly",
      "prompt_strength": 0.8,
      "num_inference_steps": 50
    }
  }' \
  http://localhost:5000/predictions

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

Output

{
  "completed_at": "2023-07-26T20:01:03.053313Z",
  "created_at": "2023-07-26T20:00:52.342022Z",
  "data_removed": false,
  "error": null,
  "id": "lrmn5btbq2o5xpbksl56fw4y64",
  "input": {
    "width": 1248,
    "height": 832,
    "prompt": "A beautiful landscape photo",
    "refine": "expert_ensemble_refiner",
    "scheduler": "KarrasDPM",
    "num_outputs": 1,
    "guidance_scale": 7.5,
    "high_noise_frac": 0.8,
    "negative_prompt": "soft, blurry, ugly",
    "prompt_strength": 0.8,
    "num_inference_steps": 50
  },
  "logs": "Using seed: 58684\ntxt2img mode\n  0%|          | 0/31 [00:00<?, ?it/s]\n  3%|▎         | 1/31 [00:00<00:06,  4.80it/s]\n  6%|▋         | 2/31 [00:00<00:05,  4.89it/s]\n 10%|▉         | 3/31 [00:00<00:05,  4.92it/s]\n 13%|█▎        | 4/31 [00:00<00:05,  4.94it/s]\n 16%|█▌        | 5/31 [00:01<00:05,  4.95it/s]\n 19%|█▉        | 6/31 [00:01<00:05,  4.95it/s]\n 23%|██▎       | 7/31 [00:01<00:04,  4.97it/s]\n 26%|██▌       | 8/31 [00:01<00:04,  4.97it/s]\n 29%|██▉       | 9/31 [00:01<00:04,  4.98it/s]\n 32%|███▏      | 10/31 [00:02<00:04,  4.98it/s]\n 35%|███▌      | 11/31 [00:02<00:04,  4.97it/s]\n 39%|███▊      | 12/31 [00:02<00:03,  4.98it/s]\n 42%|████▏     | 13/31 [00:02<00:03,  4.98it/s]\n 45%|████▌     | 14/31 [00:02<00:03,  4.99it/s]\n 48%|████▊     | 15/31 [00:03<00:03,  4.98it/s]\n 52%|█████▏    | 16/31 [00:03<00:03,  4.98it/s]\n 55%|█████▍    | 17/31 [00:03<00:02,  4.98it/s]\n 58%|█████▊    | 18/31 [00:03<00:02,  4.98it/s]\n 61%|██████▏   | 19/31 [00:03<00:02,  4.98it/s]\n 65%|██████▍   | 20/31 [00:04<00:02,  4.97it/s]\n 68%|██████▊   | 21/31 [00:04<00:02,  4.98it/s]\n 71%|███████   | 22/31 [00:04<00:01,  4.97it/s]\n 74%|███████▍  | 23/31 [00:04<00:01,  4.97it/s]\n 77%|███████▋  | 24/31 [00:04<00:01,  4.97it/s]\n 81%|████████  | 25/31 [00:05<00:01,  4.97it/s]\n 84%|████████▍ | 26/31 [00:05<00:01,  4.97it/s]\n 87%|████████▋ | 27/31 [00:05<00:00,  4.97it/s]\n 90%|█████████ | 28/31 [00:05<00:00,  4.97it/s]\n 94%|█████████▎| 29/31 [00:05<00:00,  4.97it/s]\n 97%|█████████▋| 30/31 [00:06<00:00,  4.97it/s]\n100%|██████████| 31/31 [00:06<00:00,  4.97it/s]\n100%|██████████| 31/31 [00:06<00:00,  4.97it/s]\n  0%|          | 0/10 [00:00<?, ?it/s]\n 10%|█         | 1/10 [00:00<00:02,  4.15it/s]\n 20%|██        | 2/10 [00:00<00:01,  4.21it/s]\n 30%|███       | 3/10 [00:00<00:01,  4.23it/s]\n 40%|████      | 4/10 [00:00<00:01,  4.24it/s]\n 50%|█████     | 5/10 [00:01<00:01,  4.24it/s]\n 60%|██████    | 6/10 [00:01<00:00,  4.24it/s]\n 70%|███████   | 7/10 [00:01<00:00,  4.24it/s]\n 80%|████████  | 8/10 [00:01<00:00,  4.24it/s]\n 90%|█████████ | 9/10 [00:02<00:00,  4.24it/s]\n100%|██████████| 10/10 [00:02<00:00,  4.24it/s]\n100%|██████████| 10/10 [00:02<00:00,  4.24it/s]",
  "metrics": {
    "predict_time": 10.725354,
    "total_time": 10.711291
  },
  "output": [
    "https://replicate.delivery/pbxt/zIPS4uyGONKvKBg9iTA6FRC785eK7eWhpewpR7W0RnF9rlniA/out-0.png"
  ],
  "started_at": "2023-07-26T20:00:52.327959Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/lrmn5btbq2o5xpbksl56fw4y64",
    "cancel": "https://api.replicate.com/v1/predictions/lrmn5btbq2o5xpbksl56fw4y64/cancel"
  },
  "version": "610dddf033f10431b1b55f24510b6009fcba23017ee551a1b9afbc4eec79e29c"
}

Generated in

10.7 seconds

Tweak itReport

Using seed: 58684
txt2img mode
  0%|          | 0/31 [00:00<?, ?it/s]
  3%|▎         | 1/31 [00:00<00:06,  4.80it/s]
  6%|▋         | 2/31 [00:00<00:05,  4.89it/s]
 10%|▉         | 3/31 [00:00<00:05,  4.92it/s]
 13%|█▎        | 4/31 [00:00<00:05,  4.94it/s]
 16%|█▌        | 5/31 [00:01<00:05,  4.95it/s]
 19%|█▉        | 6/31 [00:01<00:05,  4.95it/s]
 23%|██▎       | 7/31 [00:01<00:04,  4.97it/s]
 26%|██▌       | 8/31 [00:01<00:04,  4.97it/s]
 29%|██▉       | 9/31 [00:01<00:04,  4.98it/s]
 32%|███▏      | 10/31 [00:02<00:04,  4.98it/s]
 35%|███▌      | 11/31 [00:02<00:04,  4.97it/s]
 39%|███▊      | 12/31 [00:02<00:03,  4.98it/s]
 42%|████▏     | 13/31 [00:02<00:03,  4.98it/s]
 45%|████▌     | 14/31 [00:02<00:03,  4.99it/s]
 48%|████▊     | 15/31 [00:03<00:03,  4.98it/s]
 52%|█████▏    | 16/31 [00:03<00:03,  4.98it/s]
 55%|█████▍    | 17/31 [00:03<00:02,  4.98it/s]
 58%|█████▊    | 18/31 [00:03<00:02,  4.98it/s]
 61%|██████▏   | 19/31 [00:03<00:02,  4.98it/s]
 65%|██████▍   | 20/31 [00:04<00:02,  4.97it/s]
 68%|██████▊   | 21/31 [00:04<00:02,  4.98it/s]
 71%|███████   | 22/31 [00:04<00:01,  4.97it/s]
 74%|███████▍  | 23/31 [00:04<00:01,  4.97it/s]
 77%|███████▋  | 24/31 [00:04<00:01,  4.97it/s]
 81%|████████  | 25/31 [00:05<00:01,  4.97it/s]
 84%|████████▍ | 26/31 [00:05<00:01,  4.97it/s]
 87%|████████▋ | 27/31 [00:05<00:00,  4.97it/s]
 90%|█████████ | 28/31 [00:05<00:00,  4.97it/s]
 94%|█████████▎| 29/31 [00:05<00:00,  4.97it/s]
 97%|█████████▋| 30/31 [00:06<00:00,  4.97it/s]
100%|██████████| 31/31 [00:06<00:00,  4.97it/s]
100%|██████████| 31/31 [00:06<00:00,  4.97it/s]
  0%|          | 0/10 [00:00<?, ?it/s]
 10%|█         | 1/10 [00:00<00:02,  4.15it/s]
 20%|██        | 2/10 [00:00<00:01,  4.21it/s]
 30%|███       | 3/10 [00:00<00:01,  4.23it/s]
 40%|████      | 4/10 [00:00<00:01,  4.24it/s]
 50%|█████     | 5/10 [00:01<00:01,  4.24it/s]
 60%|██████    | 6/10 [00:01<00:00,  4.24it/s]
 70%|███████   | 7/10 [00:01<00:00,  4.24it/s]
 80%|████████  | 8/10 [00:01<00:00,  4.24it/s]
 90%|█████████ | 9/10 [00:02<00:00,  4.24it/s]
100%|██████████| 10/10 [00:02<00:00,  4.24it/s]
100%|██████████| 10/10 [00:02<00:00,  4.24it/s]