fofr/sdxl-ms-paint – Run with an API on Replicate

fofr / sdxl-ms-paint

Make MS Paint images, a lora trained by aimingfail

Cold

Public
1.1K runs
L40S
Paper

Run with an API

Playground API Examples README Versions

Input

prompt

string

Shift + Return to add a new line

Default: ""

negative_prompt

string

Shift + Return to add a new line

Things you do not want to see in your image

Default: ""

number_of_images

integer

(minimum: 1, maximum: 10)

Number of images to generate

Default: 1

width

integer

Default: 1024

height

integer

Default: 1024

lora_strength

number

(minimum: 0, maximum: 3)

Strength of the lora to use for the generation. Default is 1.0.

Default: 1

output_format

string

Format of the output images

Default: "webp"

output_quality

integer

(minimum: 0, maximum: 100)

Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality.

Default: 80

seed

integer

Set a seed for reproducibility. Random by default.

disable_safety_checker

boolean

Disable safety checker for generated images.

Default: false

sampler

string

Advanced. Change the sampler used for generation. Default is what we think gives the best images.

Default: "Default"

scheduler

string

Advanced. Change the scheduler used for generation. Default is what we think gives the best images.

Default: "Default"

steps

integer

(minimum: 1, maximum: 50)

Advanced. Leave empty to use recommended steps. Set it only if you want to customise the number of steps to run the sampler for.

cfg

number

(minimum: 0, maximum: 20)

Advanced. Leave empty to use recommended CFG (classifier free guidance). This changes how much the prompt influences the output. Set it only if you want to customise.

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

npx create-replicate --model=fofr/sdxl-ms-paint

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

const output = await replicate.run(
  "fofr/sdxl-ms-paint:8e36d00024aa9a55c7d8ef0f82ccd82ab4d3225d1e702149179498258177ca3f",
  {
    input: {
      width: 1024,
      height: 1024,
      prompt: "MSPaint portrait of a dog",
      sampler: "Default",
      scheduler: "Default",
      lora_strength: 1,
      output_format: "webp",
      output_quality: 80,
      negative_prompt: "",
      number_of_images: 1
    }
  }
);

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

output = replicate.run(
    "fofr/sdxl-ms-paint:8e36d00024aa9a55c7d8ef0f82ccd82ab4d3225d1e702149179498258177ca3f",
    input={
        "width": 1024,
        "height": 1024,
        "prompt": "MSPaint portrait of a dog",
        "sampler": "Default",
        "scheduler": "Default",
        "lora_strength": 1,
        "output_format": "webp",
        "output_quality": 80,
        "negative_prompt": "",
        "number_of_images": 1
    }
)
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 fofr/sdxl-ms-paint 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": "fofr/sdxl-ms-paint:8e36d00024aa9a55c7d8ef0f82ccd82ab4d3225d1e702149179498258177ca3f",
    "input": {
      "width": 1024,
      "height": 1024,
      "prompt": "MSPaint portrait of a dog",
      "sampler": "Default",
      "scheduler": "Default",
      "lora_strength": 1,
      "output_format": "webp",
      "output_quality": 80,
      "negative_prompt": "",
      "number_of_images": 1
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2024-06-08T22:29:41.930129Z",
  "created_at": "2024-06-08T22:28:25.632000Z",
  "data_removed": false,
  "error": null,
  "id": "dq5z1emxm1rgj0cfz7vvskbt14",
  "input": {
    "width": 1024,
    "height": 1024,
    "prompt": "MSPaint portrait of a dog",
    "sampler": "Default",
    "scheduler": "Default",
    "lora_strength": 1,
    "output_format": "webp",
    "output_quality": 80,
    "negative_prompt": "",
    "number_of_images": 1
  },
  "logs": "Random seed set to: 2783082920\n==============================\nGeneration settings\nSampler: dpmpp_2m_sde_gpu\nScheduler: karras\nSteps: 20\nCFG: 7.0\nLORA: Using a lora. Lora strength: 1.0\n==============================\nRunning workflow\ngot prompt\nExecuting node 4, title: Load Checkpoint, class type: CheckpointLoaderSimple\nmodel_type EPS\nUsing pytorch attention in VAE\nUsing pytorch attention in VAE\nclip missing: ['clip_l.logit_scale', 'clip_l.transformer.text_projection.weight']\nloaded straight to GPU\nRequested to load SDXL\nLoading 1 new model\nExecuting node 10, title: Load LoRA, class type: LoraLoader\nExecuting node 6, title: CLIP Text Encode (Prompt), class type: CLIPTextEncode\nRequested to load SDXLClipModel\nLoading 1 new model\nExecuting node 7, title: CLIP Text Encode (Prompt), class type: CLIPTextEncode\nExecuting node 5, title: Empty Latent Image, class type: EmptyLatentImage\nExecuting node 3, title: KSampler, class type: KSampler\nRequested to load SDXL\nLoading 1 new model\n  0%|          | 0/20 [00:00<?, ?it/s]/root/.pyenv/versions/3.10.6/lib/python3.10/site-packages/torchsde/_brownian/brownian_interval.py:608: UserWarning: Should have tb<=t1 but got tb=14.614643096923828 and t1=14.614643.\nwarnings.warn(f\"Should have {tb_name}<=t1 but got {tb_name}={tb} and t1={self._end}.\")\n  5%|▌         | 1/20 [00:00<00:11,  1.61it/s]\n 10%|█         | 2/20 [00:00<00:06,  2.70it/s]\n 15%|█▌        | 3/20 [00:01<00:04,  3.46it/s]\n 20%|██        | 4/20 [00:01<00:04,  3.99it/s]\n 25%|██▌       | 5/20 [00:01<00:03,  4.36it/s]\n 30%|███       | 6/20 [00:01<00:03,  4.61it/s]\n 35%|███▌      | 7/20 [00:01<00:02,  4.79it/s]\n 40%|████      | 8/20 [00:01<00:02,  4.90it/s]\n 45%|████▌     | 9/20 [00:02<00:02,  4.99it/s]\n 50%|█████     | 10/20 [00:02<00:01,  5.05it/s]\n 55%|█████▌    | 11/20 [00:02<00:01,  5.10it/s]\n 60%|██████    | 12/20 [00:02<00:01,  5.14it/s]\n 65%|██████▌   | 13/20 [00:02<00:01,  5.15it/s]\n 70%|███████   | 14/20 [00:03<00:01,  5.17it/s]\n 75%|███████▌  | 15/20 [00:03<00:00,  5.17it/s]\n 80%|████████  | 16/20 [00:03<00:00,  5.18it/s]\n 85%|████████▌ | 17/20 [00:03<00:00,  5.19it/s]\n 90%|█████████ | 18/20 [00:03<00:00,  5.20it/s]\n 95%|█████████▌| 19/20 [00:04<00:00,  5.24it/s]\n100%|██████████| 20/20 [00:04<00:00,  5.26it/s]\n100%|██████████| 20/20 [00:04<00:00,  4.69it/s]\nRequested to load AutoencoderKL\nLoading 1 new model\nExecuting node 8, title: VAE Decode, class type: VAEDecode\nExecuting node 9, title: Save Image, class type: SaveImage\nPrompt executed in 8.21 seconds\noutputs:  {'9': {'images': [{'filename': 'R8__00001_.png', 'subfolder': '', 'type': 'output'}]}}\n====================================\nR8__00001_.png",
  "metrics": {
    "predict_time": 10.870445,
    "total_time": 76.298129
  },
  "output": [
    "https://replicate.delivery/pbxt/Kf3ix6vdRp0xbiR5139W1CeUtLzwJfyMAiyilmGNcrsqqR5lA/R8__00001_.webp"
  ],
  "started_at": "2024-06-08T22:29:31.059684Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/dq5z1emxm1rgj0cfz7vvskbt14",
    "cancel": "https://api.replicate.com/v1/predictions/dq5z1emxm1rgj0cfz7vvskbt14/cancel"
  },
  "version": "8e36d00024aa9a55c7d8ef0f82ccd82ab4d3225d1e702149179498258177ca3f"
}

Generated in

10.9 seconds

Tweak it ShareReport

Random seed set to: 2783082920
==============================
Generation settings
Sampler: dpmpp_2m_sde_gpu
Scheduler: karras
Steps: 20
CFG: 7.0
LORA: Using a lora. Lora strength: 1.0
==============================
Running workflow
got prompt
Executing node 4, title: Load Checkpoint, class type: CheckpointLoaderSimple
model_type EPS
Using pytorch attention in VAE
Using pytorch attention in VAE
clip missing: ['clip_l.logit_scale', 'clip_l.transformer.text_projection.weight']
loaded straight to GPU
Requested to load SDXL
Loading 1 new model
Executing node 10, title: Load LoRA, class type: LoraLoader
Executing node 6, title: CLIP Text Encode (Prompt), class type: CLIPTextEncode
Requested to load SDXLClipModel
Loading 1 new model
Executing node 7, title: CLIP Text Encode (Prompt), class type: CLIPTextEncode
Executing node 5, title: Empty Latent Image, class type: EmptyLatentImage
Executing node 3, title: KSampler, class type: KSampler
Requested to load SDXL
Loading 1 new model
  0%|          | 0/20 [00:00<?, ?it/s]/root/.pyenv/versions/3.10.6/lib/python3.10/site-packages/torchsde/_brownian/brownian_interval.py:608: UserWarning: Should have tb<=t1 but got tb=14.614643096923828 and t1=14.614643.
warnings.warn(f"Should have {tb_name}<=t1 but got {tb_name}={tb} and t1={self._end}.")
  5%|▌         | 1/20 [00:00<00:11,  1.61it/s]
 10%|█         | 2/20 [00:00<00:06,  2.70it/s]
 15%|█▌        | 3/20 [00:01<00:04,  3.46it/s]
 20%|██        | 4/20 [00:01<00:04,  3.99it/s]
 25%|██▌       | 5/20 [00:01<00:03,  4.36it/s]
 30%|███       | 6/20 [00:01<00:03,  4.61it/s]
 35%|███▌      | 7/20 [00:01<00:02,  4.79it/s]
 40%|████      | 8/20 [00:01<00:02,  4.90it/s]
 45%|████▌     | 9/20 [00:02<00:02,  4.99it/s]
 50%|█████     | 10/20 [00:02<00:01,  5.05it/s]
 55%|█████▌    | 11/20 [00:02<00:01,  5.10it/s]
 60%|██████    | 12/20 [00:02<00:01,  5.14it/s]
 65%|██████▌   | 13/20 [00:02<00:01,  5.15it/s]
 70%|███████   | 14/20 [00:03<00:01,  5.17it/s]
 75%|███████▌  | 15/20 [00:03<00:00,  5.17it/s]
 80%|████████  | 16/20 [00:03<00:00,  5.18it/s]
 85%|████████▌ | 17/20 [00:03<00:00,  5.19it/s]
 90%|█████████ | 18/20 [00:03<00:00,  5.20it/s]
 95%|█████████▌| 19/20 [00:04<00:00,  5.24it/s]
100%|██████████| 20/20 [00:04<00:00,  5.26it/s]
100%|██████████| 20/20 [00:04<00:00,  4.69it/s]
Requested to load AutoencoderKL
Loading 1 new model
Executing node 8, title: VAE Decode, class type: VAEDecode
Executing node 9, title: Save Image, class type: SaveImage
Prompt executed in 8.21 seconds
outputs:  {'9': {'images': [{'filename': 'R8__00001_.png', 'subfolder': '', 'type': 'output'}]}}
====================================
R8__00001_.png

Examples

View more examples

Run time and cost

This model costs approximately $0.0028 to run on Replicate, or 357 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 3 seconds.

Readme

Just prompt: MSPaint portrait of {subject} Or: MSPaint drawing of {scene}

https://civitai.com/models/183354/sdxl-ms-paint-portraits