roblester/editorial-cartoon – Run with an API on Replicate

roblester / editorial-cartoon

An SDXL model Finetuned for style on a bespoke dataset of Stylized Illustrations

Warm

Public
1.3K runs
L40S
SDXL fine-tune

Run with an API

Playground API Examples README Versions

Input

prompt

string

Shift + Return to add a new line

illustration in the style of TOK of A nurse on the street in NYCillustration in the style of TOK of A nurse on the street in NYC

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. 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: "K_EULER"

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

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

lora_scale

number

(minimum: 0, maximum: 1)

LoRA additive scale. Only applicable on trained models.

Default: 0.6

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](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=roblester/editorial-cartoon

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 roblester/editorial-cartoon using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "roblester/editorial-cartoon:873e1db54db20146e5bb885a9e6e1dc268a62845d7fbfd3863f6507c938d844b",
  {
    input: {
      width: 1024,
      height: 1024,
      prompt: "illustration in the style of TOK of A nurse on the street in NYC",
      refine: "expert_ensemble_refiner",
      scheduler: "K_EULER_ANCESTRAL",
      lora_scale: 0.8,
      num_outputs: 1,
      guidance_scale: 7.5,
      apply_watermark: true,
      high_noise_frac: 0.8,
      negative_prompt: "",
      prompt_strength: 0.8,
      num_inference_steps: 100
    }
  }
);

// 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 roblester/editorial-cartoon using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "roblester/editorial-cartoon:873e1db54db20146e5bb885a9e6e1dc268a62845d7fbfd3863f6507c938d844b",
    input={
        "width": 1024,
        "height": 1024,
        "prompt": "illustration in the style of TOK of A nurse on the street in NYC",
        "refine": "expert_ensemble_refiner",
        "scheduler": "K_EULER_ANCESTRAL",
        "lora_scale": 0.8,
        "num_outputs": 1,
        "guidance_scale": 7.5,
        "apply_watermark": True,
        "high_noise_frac": 0.8,
        "negative_prompt": "",
        "prompt_strength": 0.8,
        "num_inference_steps": 100
    }
)
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 roblester/editorial-cartoon 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": "873e1db54db20146e5bb885a9e6e1dc268a62845d7fbfd3863f6507c938d844b",
    "input": {
      "width": 1024,
      "height": 1024,
      "prompt": "illustration in the style of TOK of A nurse on the street in NYC",
      "refine": "expert_ensemble_refiner",
      "scheduler": "K_EULER_ANCESTRAL",
      "lora_scale": 0.8,
      "num_outputs": 1,
      "guidance_scale": 7.5,
      "apply_watermark": true,
      "high_noise_frac": 0.8,
      "negative_prompt": "",
      "prompt_strength": 0.8,
      "num_inference_steps": 100
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2024-07-24T20:14:59.780809Z",
  "created_at": "2024-07-24T20:14:26.910000Z",
  "data_removed": false,
  "error": null,
  "id": "k657gyvwbsrgj0cgwsgv7ytmaw",
  "input": {
    "width": 1024,
    "height": 1024,
    "prompt": "illustration in the style of TOK of A nurse on the street in NYC",
    "refine": "expert_ensemble_refiner",
    "scheduler": "K_EULER_ANCESTRAL",
    "lora_scale": 0.8,
    "num_outputs": 1,
    "guidance_scale": 7.5,
    "apply_watermark": true,
    "high_noise_frac": 0.8,
    "negative_prompt": "",
    "prompt_strength": 0.8,
    "num_inference_steps": 100
  },
  "logs": "Using seed: 6691\nEnsuring enough disk space...\nFree disk space: 1609035563008\nDownloading weights: https://replicate.delivery/pbxt/EsFWqRfmVkzlS6RuZMMAbUbGaB7DNYYYLCl0zmFX3LXCJ4lJA/trained_model.tar\n2024-07-24T20:14:29Z | INFO  | [ Initiating ] chunk_size=150M dest=/src/weights-cache/b3834eb6ecebe924 url=https://replicate.delivery/pbxt/EsFWqRfmVkzlS6RuZMMAbUbGaB7DNYYYLCl0zmFX3LXCJ4lJA/trained_model.tar\n2024-07-24T20:14:32Z | INFO  | [ Complete ] dest=/src/weights-cache/b3834eb6ecebe924 size=\"186 MB\" total_elapsed=2.598s url=https://replicate.delivery/pbxt/EsFWqRfmVkzlS6RuZMMAbUbGaB7DNYYYLCl0zmFX3LXCJ4lJA/trained_model.tar\nb''\nDownloaded weights in 2.7233593463897705 seconds\nLoading fine-tuned model\nDoes not have Unet. assume we are using LoRA\nLoading Unet LoRA\nPrompt: illustration in the style of <s0><s1> of A nurse on the street in NYC\ntxt2img mode\n  0%|          | 0/80 [00:00<?, ?it/s]/usr/local/lib/python3.9/site-packages/torch/nn/modules/conv.py:459: UserWarning: Applied workaround for CuDNN issue, install nvrtc.so (Triggered internally at ../aten/src/ATen/native/cudnn/Conv_v8.cpp:80.)\nreturn F.conv2d(input, weight, bias, self.stride,\n/usr/local/lib/python3.9/site-packages/diffusers/models/attention_processor.py:1946: FutureWarning: `LoRAAttnProcessor2_0` is deprecated and will be removed in version 0.26.0. Make sure use AttnProcessor2_0 instead by settingLoRA layers to `self.{to_q,to_k,to_v,to_out[0]}.lora_layer` respectively. This will be done automatically when using `LoraLoaderMixin.load_lora_weights`\ndeprecate(\n  1%|▏         | 1/80 [00:00<00:33,  2.36it/s]\n  2%|▎         | 2/80 [00:00<00:20,  3.72it/s]\n  4%|▍         | 3/80 [00:00<00:19,  3.95it/s]\n  5%|▌         | 4/80 [00:01<00:18,  4.07it/s]\n  6%|▋         | 5/80 [00:01<00:18,  4.13it/s]\n  8%|▊         | 6/80 [00:01<00:17,  4.16it/s]\n  9%|▉         | 7/80 [00:01<00:17,  4.19it/s]\n 10%|█         | 8/80 [00:01<00:17,  4.21it/s]\n 11%|█▏        | 9/80 [00:02<00:16,  4.21it/s]\n 12%|█▎        | 10/80 [00:02<00:16,  4.21it/s]\n 14%|█▍        | 11/80 [00:02<00:16,  4.22it/s]\n 15%|█▌        | 12/80 [00:02<00:16,  4.22it/s]\n 16%|█▋        | 13/80 [00:03<00:15,  4.23it/s]\n 18%|█▊        | 14/80 [00:03<00:15,  4.23it/s]\n 19%|█▉        | 15/80 [00:03<00:15,  4.22it/s]\n 20%|██        | 16/80 [00:03<00:15,  4.22it/s]\n 21%|██▏       | 17/80 [00:04<00:14,  4.23it/s]\n 22%|██▎       | 18/80 [00:04<00:14,  4.22it/s]\n 24%|██▍       | 19/80 [00:04<00:14,  4.23it/s]\n 25%|██▌       | 20/80 [00:04<00:14,  4.24it/s]\n 26%|██▋       | 21/80 [00:05<00:13,  4.24it/s]\n 28%|██▊       | 22/80 [00:05<00:13,  4.25it/s]\n 29%|██▉       | 23/80 [00:05<00:13,  4.25it/s]\n 30%|███       | 24/80 [00:05<00:13,  4.25it/s]\n 31%|███▏      | 25/80 [00:06<00:12,  4.25it/s]\n 32%|███▎      | 26/80 [00:06<00:12,  4.25it/s]\n 34%|███▍      | 27/80 [00:06<00:12,  4.25it/s]\n 35%|███▌      | 28/80 [00:06<00:12,  4.26it/s]\n 36%|███▋      | 29/80 [00:06<00:11,  4.25it/s]\n 38%|███▊      | 30/80 [00:07<00:11,  4.25it/s]\n 39%|███▉      | 31/80 [00:07<00:11,  4.25it/s]\n 40%|████      | 32/80 [00:07<00:11,  4.25it/s]\n 41%|████▏     | 33/80 [00:07<00:11,  4.25it/s]\n 42%|████▎     | 34/80 [00:08<00:10,  4.25it/s]\n 44%|████▍     | 35/80 [00:08<00:10,  4.25it/s]\n 45%|████▌     | 36/80 [00:08<00:10,  4.25it/s]\n 46%|████▋     | 37/80 [00:08<00:10,  4.24it/s]\n 48%|████▊     | 38/80 [00:09<00:09,  4.24it/s]\n 49%|████▉     | 39/80 [00:09<00:09,  4.24it/s]\n 50%|█████     | 40/80 [00:09<00:09,  4.24it/s]\n 51%|█████▏    | 41/80 [00:09<00:09,  4.24it/s]\n 52%|█████▎    | 42/80 [00:10<00:08,  4.24it/s]\n 54%|█████▍    | 43/80 [00:10<00:08,  4.24it/s]\n 55%|█████▌    | 44/80 [00:10<00:08,  4.24it/s]\n 56%|█████▋    | 45/80 [00:10<00:08,  4.24it/s]\n 57%|█████▊    | 46/80 [00:10<00:08,  4.25it/s]\n 59%|█████▉    | 47/80 [00:11<00:07,  4.25it/s]\n 60%|██████    | 48/80 [00:11<00:07,  4.25it/s]\n 61%|██████▏   | 49/80 [00:11<00:07,  4.24it/s]\n 62%|██████▎   | 50/80 [00:11<00:07,  4.24it/s]\n 64%|██████▍   | 51/80 [00:12<00:06,  4.24it/s]\n 65%|██████▌   | 52/80 [00:12<00:06,  4.24it/s]\n 66%|██████▋   | 53/80 [00:12<00:06,  4.24it/s]\n 68%|██████▊   | 54/80 [00:12<00:06,  4.24it/s]\n 69%|██████▉   | 55/80 [00:13<00:05,  4.24it/s]\n 70%|███████   | 56/80 [00:13<00:05,  4.24it/s]\n 71%|███████▏  | 57/80 [00:13<00:05,  4.24it/s]\n 72%|███████▎  | 58/80 [00:13<00:05,  4.24it/s]\n 74%|███████▍  | 59/80 [00:14<00:04,  4.24it/s]\n 75%|███████▌  | 60/80 [00:14<00:04,  4.24it/s]\n 76%|███████▋  | 61/80 [00:14<00:04,  4.24it/s]\n 78%|███████▊  | 62/80 [00:14<00:04,  4.23it/s]\n 79%|███████▉  | 63/80 [00:14<00:04,  4.23it/s]\n 80%|████████  | 64/80 [00:15<00:03,  4.23it/s]\n 81%|████████▏ | 65/80 [00:15<00:03,  4.23it/s]\n 82%|████████▎ | 66/80 [00:15<00:03,  4.23it/s]\n 84%|████████▍ | 67/80 [00:15<00:03,  4.22it/s]\n 85%|████████▌ | 68/80 [00:16<00:02,  4.23it/s]\n 86%|████████▋ | 69/80 [00:16<00:02,  4.22it/s]\n 88%|████████▊ | 70/80 [00:16<00:02,  4.22it/s]\n 89%|████████▉ | 71/80 [00:16<00:02,  4.22it/s]\n 90%|█████████ | 72/80 [00:17<00:01,  4.22it/s]\n 91%|█████████▏| 73/80 [00:17<00:01,  4.22it/s]\n 92%|█████████▎| 74/80 [00:17<00:01,  4.22it/s]\n 94%|█████████▍| 75/80 [00:17<00:01,  4.22it/s]\n 95%|█████████▌| 76/80 [00:18<00:00,  4.22it/s]\n 96%|█████████▋| 77/80 [00:18<00:00,  4.22it/s]\n 98%|█████████▊| 78/80 [00:18<00:00,  4.22it/s]\n 99%|█████████▉| 79/80 [00:18<00:00,  4.22it/s]\n100%|██████████| 80/80 [00:18<00:00,  4.22it/s]\n100%|██████████| 80/80 [00:18<00:00,  4.21it/s]\n  0%|          | 0/20 [00:00<?, ?it/s]\n  5%|▌         | 1/20 [00:00<00:05,  3.66it/s]\n 10%|█         | 2/20 [00:00<00:04,  3.96it/s]\n 15%|█▌        | 3/20 [00:00<00:04,  4.07it/s]\n 20%|██        | 4/20 [00:00<00:03,  4.12it/s]\n 25%|██▌       | 5/20 [00:01<00:03,  4.14it/s]\n 30%|███       | 6/20 [00:01<00:03,  4.16it/s]\n 35%|███▌      | 7/20 [00:01<00:03,  4.18it/s]\n 40%|████      | 8/20 [00:01<00:02,  4.19it/s]\n 45%|████▌     | 9/20 [00:02<00:02,  4.20it/s]\n 50%|█████     | 10/20 [00:02<00:02,  4.20it/s]\n 55%|█████▌    | 11/20 [00:02<00:02,  4.20it/s]\n 60%|██████    | 12/20 [00:02<00:01,  4.20it/s]\n 65%|██████▌   | 13/20 [00:03<00:01,  4.20it/s]\n 70%|███████   | 14/20 [00:03<00:01,  4.20it/s]\n 75%|███████▌  | 15/20 [00:03<00:01,  4.20it/s]\n 80%|████████  | 16/20 [00:03<00:00,  4.20it/s]\n 85%|████████▌ | 17/20 [00:04<00:00,  4.20it/s]\n 90%|█████████ | 18/20 [00:04<00:00,  4.20it/s]\n 95%|█████████▌| 19/20 [00:04<00:00,  4.20it/s]\n100%|██████████| 20/20 [00:04<00:00,  4.21it/s]\n100%|██████████| 20/20 [00:04<00:00,  4.17it/s]",
  "metrics": {
    "predict_time": 30.088841679,
    "total_time": 32.870809
  },
  "output": [
    "https://replicate.delivery/pbxt/uEcLP6VZlwpXCVKp4hAKaq9oJaFslBRluJ0fpXUvm0Shl4lJA/out-0.png"
  ],
  "started_at": "2024-07-24T20:14:29.691967Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/k657gyvwbsrgj0cgwsgv7ytmaw",
    "cancel": "https://api.replicate.com/v1/predictions/k657gyvwbsrgj0cgwsgv7ytmaw/cancel"
  },
  "version": "873e1db54db20146e5bb885a9e6e1dc268a62845d7fbfd3863f6507c938d844b"
}

Generated in

30.1 seconds

Tweak it Share Report

Using seed: 6691
Ensuring enough disk space...
Free disk space: 1609035563008
Downloading weights: https://replicate.delivery/pbxt/EsFWqRfmVkzlS6RuZMMAbUbGaB7DNYYYLCl0zmFX3LXCJ4lJA/trained_model.tar
2024-07-24T20:14:29Z | INFO  | [ Initiating ] chunk_size=150M dest=/src/weights-cache/b3834eb6ecebe924 url=https://replicate.delivery/pbxt/EsFWqRfmVkzlS6RuZMMAbUbGaB7DNYYYLCl0zmFX3LXCJ4lJA/trained_model.tar
2024-07-24T20:14:32Z | INFO  | [ Complete ] dest=/src/weights-cache/b3834eb6ecebe924 size="186 MB" total_elapsed=2.598s url=https://replicate.delivery/pbxt/EsFWqRfmVkzlS6RuZMMAbUbGaB7DNYYYLCl0zmFX3LXCJ4lJA/trained_model.tar
b''
Downloaded weights in 2.7233593463897705 seconds
Loading fine-tuned model
Does not have Unet. assume we are using LoRA
Loading Unet LoRA
Prompt: illustration in the style of <s0><s1> of A nurse on the street in NYC
txt2img mode
  0%|          | 0/80 [00:00<?, ?it/s]/usr/local/lib/python3.9/site-packages/torch/nn/modules/conv.py:459: UserWarning: Applied workaround for CuDNN issue, install nvrtc.so (Triggered internally at ../aten/src/ATen/native/cudnn/Conv_v8.cpp:80.)
return F.conv2d(input, weight, bias, self.stride,
/usr/local/lib/python3.9/site-packages/diffusers/models/attention_processor.py:1946: FutureWarning: `LoRAAttnProcessor2_0` is deprecated and will be removed in version 0.26.0. Make sure use AttnProcessor2_0 instead by settingLoRA layers to `self.{to_q,to_k,to_v,to_out[0]}.lora_layer` respectively. This will be done automatically when using `LoraLoaderMixin.load_lora_weights`
deprecate(
  1%|▏         | 1/80 [00:00<00:33,  2.36it/s]
  2%|▎         | 2/80 [00:00<00:20,  3.72it/s]
  4%|▍         | 3/80 [00:00<00:19,  3.95it/s]
  5%|▌         | 4/80 [00:01<00:18,  4.07it/s]
  6%|▋         | 5/80 [00:01<00:18,  4.13it/s]
  8%|▊         | 6/80 [00:01<00:17,  4.16it/s]
  9%|▉         | 7/80 [00:01<00:17,  4.19it/s]
 10%|█         | 8/80 [00:01<00:17,  4.21it/s]
 11%|█▏        | 9/80 [00:02<00:16,  4.21it/s]
 12%|█▎        | 10/80 [00:02<00:16,  4.21it/s]
 14%|█▍        | 11/80 [00:02<00:16,  4.22it/s]
 15%|█▌        | 12/80 [00:02<00:16,  4.22it/s]
 16%|█▋        | 13/80 [00:03<00:15,  4.23it/s]
 18%|█▊        | 14/80 [00:03<00:15,  4.23it/s]
 19%|█▉        | 15/80 [00:03<00:15,  4.22it/s]
 20%|██        | 16/80 [00:03<00:15,  4.22it/s]
 21%|██▏       | 17/80 [00:04<00:14,  4.23it/s]
 22%|██▎       | 18/80 [00:04<00:14,  4.22it/s]
 24%|██▍       | 19/80 [00:04<00:14,  4.23it/s]
 25%|██▌       | 20/80 [00:04<00:14,  4.24it/s]
 26%|██▋       | 21/80 [00:05<00:13,  4.24it/s]
 28%|██▊       | 22/80 [00:05<00:13,  4.25it/s]
 29%|██▉       | 23/80 [00:05<00:13,  4.25it/s]
 30%|███       | 24/80 [00:05<00:13,  4.25it/s]
 31%|███▏      | 25/80 [00:06<00:12,  4.25it/s]
 32%|███▎      | 26/80 [00:06<00:12,  4.25it/s]
 34%|███▍      | 27/80 [00:06<00:12,  4.25it/s]
 35%|███▌      | 28/80 [00:06<00:12,  4.26it/s]
 36%|███▋      | 29/80 [00:06<00:11,  4.25it/s]
 38%|███▊      | 30/80 [00:07<00:11,  4.25it/s]
 39%|███▉      | 31/80 [00:07<00:11,  4.25it/s]
 40%|████      | 32/80 [00:07<00:11,  4.25it/s]
 41%|████▏     | 33/80 [00:07<00:11,  4.25it/s]
 42%|████▎     | 34/80 [00:08<00:10,  4.25it/s]
 44%|████▍     | 35/80 [00:08<00:10,  4.25it/s]
 45%|████▌     | 36/80 [00:08<00:10,  4.25it/s]
 46%|████▋     | 37/80 [00:08<00:10,  4.24it/s]
 48%|████▊     | 38/80 [00:09<00:09,  4.24it/s]
 49%|████▉     | 39/80 [00:09<00:09,  4.24it/s]
 50%|█████     | 40/80 [00:09<00:09,  4.24it/s]
 51%|█████▏    | 41/80 [00:09<00:09,  4.24it/s]
 52%|█████▎    | 42/80 [00:10<00:08,  4.24it/s]
 54%|█████▍    | 43/80 [00:10<00:08,  4.24it/s]
 55%|█████▌    | 44/80 [00:10<00:08,  4.24it/s]
 56%|█████▋    | 45/80 [00:10<00:08,  4.24it/s]
 57%|█████▊    | 46/80 [00:10<00:08,  4.25it/s]
 59%|█████▉    | 47/80 [00:11<00:07,  4.25it/s]
 60%|██████    | 48/80 [00:11<00:07,  4.25it/s]
 61%|██████▏   | 49/80 [00:11<00:07,  4.24it/s]
 62%|██████▎   | 50/80 [00:11<00:07,  4.24it/s]
 64%|██████▍   | 51/80 [00:12<00:06,  4.24it/s]
 65%|██████▌   | 52/80 [00:12<00:06,  4.24it/s]
 66%|██████▋   | 53/80 [00:12<00:06,  4.24it/s]
 68%|██████▊   | 54/80 [00:12<00:06,  4.24it/s]
 69%|██████▉   | 55/80 [00:13<00:05,  4.24it/s]
 70%|███████   | 56/80 [00:13<00:05,  4.24it/s]
 71%|███████▏  | 57/80 [00:13<00:05,  4.24it/s]
 72%|███████▎  | 58/80 [00:13<00:05,  4.24it/s]
 74%|███████▍  | 59/80 [00:14<00:04,  4.24it/s]
 75%|███████▌  | 60/80 [00:14<00:04,  4.24it/s]
 76%|███████▋  | 61/80 [00:14<00:04,  4.24it/s]
 78%|███████▊  | 62/80 [00:14<00:04,  4.23it/s]
 79%|███████▉  | 63/80 [00:14<00:04,  4.23it/s]
 80%|████████  | 64/80 [00:15<00:03,  4.23it/s]
 81%|████████▏ | 65/80 [00:15<00:03,  4.23it/s]
 82%|████████▎ | 66/80 [00:15<00:03,  4.23it/s]
 84%|████████▍ | 67/80 [00:15<00:03,  4.22it/s]
 85%|████████▌ | 68/80 [00:16<00:02,  4.23it/s]
 86%|████████▋ | 69/80 [00:16<00:02,  4.22it/s]
 88%|████████▊ | 70/80 [00:16<00:02,  4.22it/s]
 89%|████████▉ | 71/80 [00:16<00:02,  4.22it/s]
 90%|█████████ | 72/80 [00:17<00:01,  4.22it/s]
 91%|█████████▏| 73/80 [00:17<00:01,  4.22it/s]
 92%|█████████▎| 74/80 [00:17<00:01,  4.22it/s]
 94%|█████████▍| 75/80 [00:17<00:01,  4.22it/s]
 95%|█████████▌| 76/80 [00:18<00:00,  4.22it/s]
 96%|█████████▋| 77/80 [00:18<00:00,  4.22it/s]
 98%|█████████▊| 78/80 [00:18<00:00,  4.22it/s]
 99%|█████████▉| 79/80 [00:18<00:00,  4.22it/s]
100%|██████████| 80/80 [00:18<00:00,  4.22it/s]
100%|██████████| 80/80 [00:18<00:00,  4.21it/s]
  0%|          | 0/20 [00:00<?, ?it/s]
  5%|▌         | 1/20 [00:00<00:05,  3.66it/s]
 10%|█         | 2/20 [00:00<00:04,  3.96it/s]
 15%|█▌        | 3/20 [00:00<00:04,  4.07it/s]
 20%|██        | 4/20 [00:00<00:03,  4.12it/s]
 25%|██▌       | 5/20 [00:01<00:03,  4.14it/s]
 30%|███       | 6/20 [00:01<00:03,  4.16it/s]
 35%|███▌      | 7/20 [00:01<00:03,  4.18it/s]
 40%|████      | 8/20 [00:01<00:02,  4.19it/s]
 45%|████▌     | 9/20 [00:02<00:02,  4.20it/s]
 50%|█████     | 10/20 [00:02<00:02,  4.20it/s]
 55%|█████▌    | 11/20 [00:02<00:02,  4.20it/s]
 60%|██████    | 12/20 [00:02<00:01,  4.20it/s]
 65%|██████▌   | 13/20 [00:03<00:01,  4.20it/s]
 70%|███████   | 14/20 [00:03<00:01,  4.20it/s]
 75%|███████▌  | 15/20 [00:03<00:01,  4.20it/s]
 80%|████████  | 16/20 [00:03<00:00,  4.20it/s]
 85%|████████▌ | 17/20 [00:04<00:00,  4.20it/s]
 90%|█████████ | 18/20 [00:04<00:00,  4.20it/s]
 95%|█████████▌| 19/20 [00:04<00:00,  4.20it/s]
100%|██████████| 20/20 [00:04<00:00,  4.21it/s]
100%|██████████| 20/20 [00:04<00:00,  4.17it/s]

Examples

View more examples

Run time and cost

This model costs approximately $0.030 to run on Replicate, or 33 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 31 seconds.

Readme

Usage Tips: INCLUDE “illustration in the style of TOK” in your prompt to trigger the finetuned LoRA model and get the application of the style.

Using image inputs can help ground the model with a representation of the scene you are trying to create, but you should also include a written description in the prompt. If you want the output to more closely resemble the input image, bring down the #prompt_strength value. (note: you will compromise some of the stylistic influence of the finetuned model by bringing down the prompt strength)

Finetuning done using a custom synthetic dataset of 24 images

Made by Robert Lester: https://www.robertlestercreative.com/