charlesmccarthy/animagine-xl – Run with an API on Replicate

charlesmccarthy / animagine-xl

Animagine XL 2.0 is an advanced latent text-to-image diffusion model designed to create high-resolution, detailed anime images.

Cold

Public
9.3K runs
L40S

Run with an API

Playground API Examples README Versions

Input

prompt

string

Shift + Return to add a new line

black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailedblack fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed

Input prompt

Default: "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed"

negative_prompt

string

Shift + Return to add a new line

Negative Input prompt

Default: "worst quality, low quality"

image

file

Input image for img2img or inpaint mode

mask

file

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

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: 100)

Number of denoising steps 60-70 for best detail, 35 for fast

Default: 60

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

apply_watermark

boolean

Applies a watermark to enable determining if an image is generated in downstream applications. If you have other provisions for generating or deploying images safely, you can use this to disable watermarking.

Default: true

disable_safety_checker

boolean

Disable safety checker for generated images. This feature is only available through the API. See https://replicate.com/docs/how-does-replicate-work#safety

Default: false

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

npx create-replicate --model=charlesmccarthy/animagine-xl

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 charlesmccarthy/animagine-xl using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "charlesmccarthy/animagine-xl:db29f76d40ecf86335295ca5b24ed95e6b1eca4e29239c47cfefa68f408cbf5e",
  {
    input: {
      width: 1024,
      height: 1024,
      prompt: "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed",
      scheduler: "KarrasDPM",
      num_outputs: 1,
      guidance_scale: 7.5,
      apply_watermark: true,
      negative_prompt: "worst quality, low quality",
      prompt_strength: 0.8,
      num_inference_steps: 60
    }
  }
);
console.log(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 charlesmccarthy/animagine-xl using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "charlesmccarthy/animagine-xl:db29f76d40ecf86335295ca5b24ed95e6b1eca4e29239c47cfefa68f408cbf5e",
    input={
        "width": 1024,
        "height": 1024,
        "prompt": "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed",
        "scheduler": "KarrasDPM",
        "num_outputs": 1,
        "guidance_scale": 7.5,
        "apply_watermark": True,
        "negative_prompt": "worst quality, low quality",
        "prompt_strength": 0.8,
        "num_inference_steps": 60
    }
)
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 charlesmccarthy/animagine-xl 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": "db29f76d40ecf86335295ca5b24ed95e6b1eca4e29239c47cfefa68f408cbf5e",
    "input": {
      "width": 1024,
      "height": 1024,
      "prompt": "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed",
      "scheduler": "KarrasDPM",
      "num_outputs": 1,
      "guidance_scale": 7.5,
      "apply_watermark": true,
      "negative_prompt": "worst quality, low quality",
      "prompt_strength": 0.8,
      "num_inference_steps": 60
    }
  }' \
  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/charlesmccarthy/animagine-xl@sha256:db29f76d40ecf86335295ca5b24ed95e6b1eca4e29239c47cfefa68f408cbf5e \
  -i 'width=1024' \
  -i 'height=1024' \
  -i 'prompt="black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed"' \
  -i 'scheduler="KarrasDPM"' \
  -i 'num_outputs=1' \
  -i 'guidance_scale=7.5' \
  -i 'apply_watermark=true' \
  -i 'negative_prompt="worst quality, low quality"' \
  -i 'prompt_strength=0.8' \
  -i 'num_inference_steps=60'

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/charlesmccarthy/animagine-xl@sha256:db29f76d40ecf86335295ca5b24ed95e6b1eca4e29239c47cfefa68f408cbf5e
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "width": 1024,
      "height": 1024,
      "prompt": "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed",
      "scheduler": "KarrasDPM",
      "num_outputs": 1,
      "guidance_scale": 7.5,
      "apply_watermark": true,
      "negative_prompt": "worst quality, low quality",
      "prompt_strength": 0.8,
      "num_inference_steps": 60
    }
  }' \
  http://localhost:5000/predictions

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

Output

{
  "completed_at": "2023-12-27T20:07:37.931232Z",
  "created_at": "2023-12-27T20:03:25.717568Z",
  "data_removed": false,
  "error": null,
  "id": "qdbj5slbxeiyh2dhh6m5feri4a",
  "input": {
    "width": 1024,
    "height": 1024,
    "prompt": "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed",
    "scheduler": "KarrasDPM",
    "num_outputs": 1,
    "guidance_scale": 7.5,
    "apply_watermark": true,
    "negative_prompt": "worst quality, low quality",
    "prompt_strength": 0.8,
    "num_inference_steps": 60
  },
  "logs": "Using seed: 4117415544\nPrompt: black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed\ntxt2img mode\n  0%|          | 0/60 [00:00<?, ?it/s]\n  2%|▏         | 1/60 [00:00<00:18,  3.15it/s]\n  5%|▌         | 3/60 [00:00<00:10,  5.25it/s]\n  7%|▋         | 4/60 [00:00<00:10,  5.17it/s]\n  8%|▊         | 5/60 [00:01<00:10,  5.12it/s]\n 10%|█         | 6/60 [00:01<00:10,  5.08it/s]\n 12%|█▏        | 7/60 [00:01<00:10,  5.04it/s]\n 13%|█▎        | 8/60 [00:01<00:10,  5.03it/s]\n 15%|█▌        | 9/60 [00:01<00:10,  5.02it/s]\n 17%|█▋        | 10/60 [00:02<00:09,  5.02it/s]\n 18%|█▊        | 11/60 [00:02<00:09,  5.01it/s]\n 20%|██        | 12/60 [00:02<00:09,  5.00it/s]\n 22%|██▏       | 13/60 [00:02<00:09,  4.99it/s]\n 23%|██▎       | 14/60 [00:02<00:09,  4.99it/s]\n 25%|██▌       | 15/60 [00:03<00:09,  4.99it/s]\n 27%|██▋       | 16/60 [00:03<00:08,  4.99it/s]\n 28%|██▊       | 17/60 [00:03<00:08,  4.98it/s]\n 30%|███       | 18/60 [00:03<00:08,  4.99it/s]\n 32%|███▏      | 19/60 [00:03<00:08,  4.99it/s]\n 33%|███▎      | 20/60 [00:04<00:08,  4.99it/s]\n 35%|███▌      | 21/60 [00:04<00:07,  4.98it/s]\n 37%|███▋      | 22/60 [00:04<00:07,  4.98it/s]\n 38%|███▊      | 23/60 [00:04<00:07,  4.98it/s]\n 40%|████      | 24/60 [00:04<00:07,  4.98it/s]\n 42%|████▏     | 25/60 [00:05<00:07,  4.98it/s]\n 43%|████▎     | 26/60 [00:05<00:06,  4.97it/s]\n 45%|████▌     | 27/60 [00:05<00:06,  4.97it/s]\n 47%|████▋     | 28/60 [00:05<00:06,  4.97it/s]\n 48%|████▊     | 29/60 [00:05<00:06,  4.97it/s]\n 50%|█████     | 30/60 [00:06<00:06,  4.97it/s]\n 52%|█████▏    | 31/60 [00:06<00:05,  4.96it/s]\n 53%|█████▎    | 32/60 [00:06<00:05,  4.96it/s]\n 55%|█████▌    | 33/60 [00:06<00:05,  4.96it/s]\n 57%|█████▋    | 34/60 [00:06<00:05,  4.96it/s]\n 58%|█████▊    | 35/60 [00:07<00:05,  4.96it/s]\n 60%|██████    | 36/60 [00:07<00:04,  4.96it/s]\n 62%|██████▏   | 37/60 [00:07<00:04,  4.95it/s]\n 63%|██████▎   | 38/60 [00:07<00:04,  4.95it/s]\n 65%|██████▌   | 39/60 [00:07<00:04,  4.96it/s]\n 67%|██████▋   | 40/60 [00:08<00:04,  4.96it/s]\n 68%|██████▊   | 41/60 [00:08<00:03,  4.96it/s]\n 70%|███████   | 42/60 [00:08<00:03,  4.95it/s]\n 72%|███████▏  | 43/60 [00:08<00:03,  4.95it/s]\n 73%|███████▎  | 44/60 [00:08<00:03,  4.95it/s]\n 75%|███████▌  | 45/60 [00:09<00:03,  4.95it/s]\n 77%|███████▋  | 46/60 [00:09<00:02,  4.95it/s]\n 78%|███████▊  | 47/60 [00:09<00:02,  4.95it/s]\n 80%|████████  | 48/60 [00:09<00:02,  4.95it/s]\n 82%|████████▏ | 49/60 [00:09<00:02,  4.95it/s]\n 83%|████████▎ | 50/60 [00:10<00:02,  4.95it/s]\n 85%|████████▌ | 51/60 [00:10<00:01,  4.95it/s]\n 87%|████████▋ | 52/60 [00:10<00:01,  4.95it/s]\n 88%|████████▊ | 53/60 [00:10<00:01,  4.94it/s]\n 90%|█████████ | 54/60 [00:10<00:01,  4.94it/s]\n 92%|█████████▏| 55/60 [00:11<00:01,  4.94it/s]\n 93%|█████████▎| 56/60 [00:11<00:00,  4.94it/s]\n 95%|█████████▌| 57/60 [00:11<00:00,  4.94it/s]\n 97%|█████████▋| 58/60 [00:11<00:00,  4.94it/s]\n 98%|█████████▊| 59/60 [00:11<00:00,  4.94it/s]\n100%|██████████| 60/60 [00:12<00:00,  4.94it/s]\n100%|██████████| 60/60 [00:12<00:00,  4.96it/s]",
  "metrics": {
    "predict_time": 13.656473,
    "total_time": 252.213664
  },
  "output": [
    "https://replicate.delivery/pbxt/IfEUxNYMl7VwViOVFVH9jHo13GRWVqFpKyAl5aAakSrEsRDJA/out-0.png"
  ],
  "started_at": "2023-12-27T20:07:24.274759Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/qdbj5slbxeiyh2dhh6m5feri4a",
    "cancel": "https://api.replicate.com/v1/predictions/qdbj5slbxeiyh2dhh6m5feri4a/cancel"
  },
  "version": "db29f76d40ecf86335295ca5b24ed95e6b1eca4e29239c47cfefa68f408cbf5e"
}

Generated in

13.7 seconds

Tweak itReport

Using seed: 4117415544
Prompt: black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed
txt2img mode
  0%|          | 0/60 [00:00<?, ?it/s]
  2%|▏         | 1/60 [00:00<00:18,  3.15it/s]
  5%|▌         | 3/60 [00:00<00:10,  5.25it/s]
  7%|▋         | 4/60 [00:00<00:10,  5.17it/s]
  8%|▊         | 5/60 [00:01<00:10,  5.12it/s]
 10%|█         | 6/60 [00:01<00:10,  5.08it/s]
 12%|█▏        | 7/60 [00:01<00:10,  5.04it/s]
 13%|█▎        | 8/60 [00:01<00:10,  5.03it/s]
 15%|█▌        | 9/60 [00:01<00:10,  5.02it/s]
 17%|█▋        | 10/60 [00:02<00:09,  5.02it/s]
 18%|█▊        | 11/60 [00:02<00:09,  5.01it/s]
 20%|██        | 12/60 [00:02<00:09,  5.00it/s]
 22%|██▏       | 13/60 [00:02<00:09,  4.99it/s]
 23%|██▎       | 14/60 [00:02<00:09,  4.99it/s]
 25%|██▌       | 15/60 [00:03<00:09,  4.99it/s]
 27%|██▋       | 16/60 [00:03<00:08,  4.99it/s]
 28%|██▊       | 17/60 [00:03<00:08,  4.98it/s]
 30%|███       | 18/60 [00:03<00:08,  4.99it/s]
 32%|███▏      | 19/60 [00:03<00:08,  4.99it/s]
 33%|███▎      | 20/60 [00:04<00:08,  4.99it/s]
 35%|███▌      | 21/60 [00:04<00:07,  4.98it/s]
 37%|███▋      | 22/60 [00:04<00:07,  4.98it/s]
 38%|███▊      | 23/60 [00:04<00:07,  4.98it/s]
 40%|████      | 24/60 [00:04<00:07,  4.98it/s]
 42%|████▏     | 25/60 [00:05<00:07,  4.98it/s]
 43%|████▎     | 26/60 [00:05<00:06,  4.97it/s]
 45%|████▌     | 27/60 [00:05<00:06,  4.97it/s]
 47%|████▋     | 28/60 [00:05<00:06,  4.97it/s]
 48%|████▊     | 29/60 [00:05<00:06,  4.97it/s]
 50%|█████     | 30/60 [00:06<00:06,  4.97it/s]
 52%|█████▏    | 31/60 [00:06<00:05,  4.96it/s]
 53%|█████▎    | 32/60 [00:06<00:05,  4.96it/s]
 55%|█████▌    | 33/60 [00:06<00:05,  4.96it/s]
 57%|█████▋    | 34/60 [00:06<00:05,  4.96it/s]
 58%|█████▊    | 35/60 [00:07<00:05,  4.96it/s]
 60%|██████    | 36/60 [00:07<00:04,  4.96it/s]
 62%|██████▏   | 37/60 [00:07<00:04,  4.95it/s]
 63%|██████▎   | 38/60 [00:07<00:04,  4.95it/s]
 65%|██████▌   | 39/60 [00:07<00:04,  4.96it/s]
 67%|██████▋   | 40/60 [00:08<00:04,  4.96it/s]
 68%|██████▊   | 41/60 [00:08<00:03,  4.96it/s]
 70%|███████   | 42/60 [00:08<00:03,  4.95it/s]
 72%|███████▏  | 43/60 [00:08<00:03,  4.95it/s]
 73%|███████▎  | 44/60 [00:08<00:03,  4.95it/s]
 75%|███████▌  | 45/60 [00:09<00:03,  4.95it/s]
 77%|███████▋  | 46/60 [00:09<00:02,  4.95it/s]
 78%|███████▊  | 47/60 [00:09<00:02,  4.95it/s]
 80%|████████  | 48/60 [00:09<00:02,  4.95it/s]
 82%|████████▏ | 49/60 [00:09<00:02,  4.95it/s]
 83%|████████▎ | 50/60 [00:10<00:02,  4.95it/s]
 85%|████████▌ | 51/60 [00:10<00:01,  4.95it/s]
 87%|████████▋ | 52/60 [00:10<00:01,  4.95it/s]
 88%|████████▊ | 53/60 [00:10<00:01,  4.94it/s]
 90%|█████████ | 54/60 [00:10<00:01,  4.94it/s]
 92%|█████████▏| 55/60 [00:11<00:01,  4.94it/s]
 93%|█████████▎| 56/60 [00:11<00:00,  4.94it/s]
 95%|█████████▌| 57/60 [00:11<00:00,  4.94it/s]
 97%|█████████▋| 58/60 [00:11<00:00,  4.94it/s]
 98%|█████████▊| 59/60 [00:11<00:00,  4.94it/s]
100%|██████████| 60/60 [00:12<00:00,  4.94it/s]
100%|██████████| 60/60 [00:12<00:00,  4.96it/s]

Examples

View more examples

Run time and cost

This model costs approximately $0.014 to run on Replicate, or 71 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 15 seconds.

Readme

Animagine XL 2.0 is an advanced latent text-to-image diffusion model designed to create high-resolution, detailed anime images. It’s fine-tuned from Stable Diffusion XL 1.0 using a high-quality anime-style image dataset. This model, an upgrade from Animagine XL 1.0, excels in capturing the diverse and distinct styles of anime art, offering improved image quality and aesthetics.