prakharsaxena24/masked-upscaler – Run with an API on Replicate

prakharsaxena24 / masked-upscaler

Upscaler and detailer for a selected area

Cold

Public
4.8K runs
A100 (80GB)

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

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

npx create-replicate --model=prakharsaxena24/masked-upscaler

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 prakharsaxena24/masked-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "prakharsaxena24/masked-upscaler:0e864cd4844ac63d862efd3468e4c55219066351009db73833ad67f98c5eaefb",
  {
    input: {
      mask: "https://replicate.delivery/pbxt/L293tY1UNaSlq01zA1VCCkNyv49jD4Ab3QrMau376xUON56q/inverse_image_mask.png",
      seed: 42,
      image: "https://replicate.delivery/pbxt/L293tzfx8WiFQQrLRxPMwRwMZHzi9Bs5a1mUOgwySqf77men/img5a21b4bd1c924b0ba6d04f1c75ced25d.png",
      prompt: "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>",
      scale_factor: 2,
      num_inference_steps: 20
    }
  }
);

// 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 prakharsaxena24/masked-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "prakharsaxena24/masked-upscaler:0e864cd4844ac63d862efd3468e4c55219066351009db73833ad67f98c5eaefb",
    input={
        "mask": "https://replicate.delivery/pbxt/L293tY1UNaSlq01zA1VCCkNyv49jD4Ab3QrMau376xUON56q/inverse_image_mask.png",
        "seed": 42,
        "image": "https://replicate.delivery/pbxt/L293tzfx8WiFQQrLRxPMwRwMZHzi9Bs5a1mUOgwySqf77men/img5a21b4bd1c924b0ba6d04f1c75ced25d.png",
        "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>",
        "scale_factor": 2,
        "num_inference_steps": 20
    }
)
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 prakharsaxena24/masked-upscaler 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": "prakharsaxena24/masked-upscaler:0e864cd4844ac63d862efd3468e4c55219066351009db73833ad67f98c5eaefb",
    "input": {
      "mask": "https://replicate.delivery/pbxt/L293tY1UNaSlq01zA1VCCkNyv49jD4Ab3QrMau376xUON56q/inverse_image_mask.png",
      "seed": 42,
      "image": "https://replicate.delivery/pbxt/L293tzfx8WiFQQrLRxPMwRwMZHzi9Bs5a1mUOgwySqf77men/img5a21b4bd1c924b0ba6d04f1c75ced25d.png",
      "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>",
      "scale_factor": 2,
      "num_inference_steps": 20
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2024-06-03T19:24:15.121352Z",
  "created_at": "2024-06-03T19:22:02.672000Z",
  "data_removed": false,
  "error": null,
  "id": "y9q613he61rgp0cfvy6sqcwjkc",
  "input": {
    "mask": "https://replicate.delivery/pbxt/L293tY1UNaSlq01zA1VCCkNyv49jD4Ab3QrMau376xUON56q/inverse_image_mask.png",
    "seed": 42,
    "image": "https://replicate.delivery/pbxt/L293tzfx8WiFQQrLRxPMwRwMZHzi9Bs5a1mUOgwySqf77men/img5a21b4bd1c924b0ba6d04f1c75ced25d.png",
    "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>",
    "scale_factor": 2,
    "num_inference_steps": 20
  },
  "logs": "Running prediction\nUpscaling with scale_factor:  2.0\n[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-06-03 19:24:07,997 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-06-03 19:24:07,997 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model from cache: control_v11f1e_sd15_tile\n2024-06-03 19:24:08,011 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-06-03 19:24:08,011 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 950\n2024-06-03 19:24:08,092 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 0.1022803783416748\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 118x112, Tile count: 3, Batch size: 3, Tile batches: 1 (ext: ContrlNet)\n[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling:   0%|          | 0/1 [00:00<?, ?it/s]\n  0%|          | 0/8 [00:00<?, ?it/s]\u001b[A\nTotal progress:   0%|          | 0/8 [00:00<?, ?it/s]\u001b[A\n 12%|█▎        | 1/8 [00:01<00:07,  1.01s/it]\u001b[A\nTotal progress:  25%|██▌       | 2/8 [00:00<00:00,  6.32it/s]\u001b[A\n 25%|██▌       | 2/8 [00:01<00:03,  1.66it/s]\u001b[A\nTotal progress:  38%|███▊      | 3/8 [00:00<00:01,  4.51it/s]\u001b[A\n 38%|███▊      | 3/8 [00:01<00:02,  2.13it/s]\u001b[A\nTotal progress:  50%|█████     | 4/8 [00:00<00:01,  3.91it/s]\u001b[A\n 50%|█████     | 4/8 [00:01<00:01,  2.45it/s]\u001b[A\nTotal progress:  62%|██████▎   | 5/8 [00:01<00:00,  3.63it/s]\u001b[A\n 62%|██████▎   | 5/8 [00:02<00:01,  2.68it/s]\u001b[A\nTotal progress:  75%|███████▌  | 6/8 [00:01<00:00,  3.48it/s]\u001b[A\n 75%|███████▌  | 6/8 [00:02<00:00,  2.84it/s]\u001b[A\nTotal progress:  88%|████████▊ | 7/8 [00:01<00:00,  3.38it/s]\u001b[A\n 88%|████████▊ | 7/8 [00:02<00:00,  2.96it/s]\u001b[A\n100%|██████████| 8/8 [00:03<00:00,  3.05it/s]\u001b[A\n100%|██████████| 8/8 [00:03<00:00,  2.51it/s]\nTotal progress: 100%|██████████| 8/8 [00:02<00:00,  3.33it/s]\u001b[A[Tiled VAE]: input_size: torch.Size([1, 4, 118, 250]), tile_size: 128, padding: 11\n[Tiled VAE]: split to 1x2 = 2 tiles. Optimal tile size 128x96, original tile size 128x128\n[Tiled VAE]: Fast mode enabled, estimating group norm parameters on 128 x 60 image\n[Tiled VAE]: Executing Decoder Task Queue:   0%|          | 0/246 [00:00<?, ?it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue:  50%|█████     | 124/246 [00:00<00:00, 672.41it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 246/246 [00:00<00:00, 744.19it/s]\n[Tiled VAE]: Done in 0.965s, max VRAM alloc 5375.187 MB\nTotal progress: 100%|██████████| 8/8 [00:03<00:00,  3.33it/s]\u001b[A\nTotal progress: 100%|██████████| 8/8 [00:03<00:00,  2.30it/s]\nPrediction took 8.56 seconds",
  "metrics": {
    "predict_time": 9.202251,
    "total_time": 132.449352
  },
  "output": [
    "https://replicate.delivery/pbxt/4n6tIgWqfnV3EyQRdqGfyYjkhye5sG4xyr3Ab20EQp99S51lA/42-dd0112c8-21de-11ef-b285-1e1adbf3739f.png"
  ],
  "started_at": "2024-06-03T19:24:05.919101Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/y9q613he61rgp0cfvy6sqcwjkc",
    "cancel": "https://api.replicate.com/v1/predictions/y9q613he61rgp0cfvy6sqcwjkc/cancel"
  },
  "version": "0e864cd4844ac63d862efd3468e4c55219066351009db73833ad67f98c5eaefb"
}

Generated in

9.2 seconds

Tweak it ShareReport

Running prediction
Upscaling with scale_factor:  2.0
[Tiled Diffusion] upscaling image with 4x-UltraSharp...
[Tiled Diffusion] ControlNet found, support is enabled.
2024-06-03 19:24:07,997 - ControlNet - INFO - unit_separate = False, style_align = False
2024-06-03 19:24:07,997 - ControlNet - INFO - Loading model from cache: control_v11f1e_sd15_tile
2024-06-03 19:24:08,011 - ControlNet - INFO - Using preprocessor: tile_resample
2024-06-03 19:24:08,011 - ControlNet - INFO - preprocessor resolution = 950
2024-06-03 19:24:08,092 - ControlNet - INFO - ControlNet Hooked - Time = 0.1022803783416748
MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 118x112, Tile count: 3, Batch size: 3, Tile batches: 1 (ext: ContrlNet)
[Tiled VAE]: the input size is tiny and unnecessary to tile.
MultiDiffusion Sampling:   0%|          | 0/1 [00:00<?, ?it/s]
  0%|          | 0/8 [00:00<?, ?it/s]
Total progress:   0%|          | 0/8 [00:00<?, ?it/s]
 12%|█▎        | 1/8 [00:01<00:07,  1.01s/it]
Total progress:  25%|██▌       | 2/8 [00:00<00:00,  6.32it/s]
 25%|██▌       | 2/8 [00:01<00:03,  1.66it/s]
Total progress:  38%|███▊      | 3/8 [00:00<00:01,  4.51it/s]
 38%|███▊      | 3/8 [00:01<00:02,  2.13it/s]
Total progress:  50%|█████     | 4/8 [00:00<00:01,  3.91it/s]
 50%|█████     | 4/8 [00:01<00:01,  2.45it/s]
Total progress:  62%|██████▎   | 5/8 [00:01<00:00,  3.63it/s]
 62%|██████▎   | 5/8 [00:02<00:01,  2.68it/s]
Total progress:  75%|███████▌  | 6/8 [00:01<00:00,  3.48it/s]
 75%|███████▌  | 6/8 [00:02<00:00,  2.84it/s]
Total progress:  88%|████████▊ | 7/8 [00:01<00:00,  3.38it/s]
 88%|████████▊ | 7/8 [00:02<00:00,  2.96it/s]
100%|██████████| 8/8 [00:03<00:00,  3.05it/s]
100%|██████████| 8/8 [00:03<00:00,  2.51it/s]
Total progress: 100%|██████████| 8/8 [00:02<00:00,  3.33it/s][Tiled VAE]: input_size: torch.Size([1, 4, 118, 250]), tile_size: 128, padding: 11
[Tiled VAE]: split to 1x2 = 2 tiles. Optimal tile size 128x96, original tile size 128x128
[Tiled VAE]: Fast mode enabled, estimating group norm parameters on 128 x 60 image
[Tiled VAE]: Executing Decoder Task Queue:   0%|          | 0/246 [00:00<?, ?it/s]
[Tiled VAE]: Executing Decoder Task Queue:  50%|█████     | 124/246 [00:00<00:00, 672.41it/s]
[Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 246/246 [00:00<00:00, 744.19it/s]
[Tiled VAE]: Done in 0.965s, max VRAM alloc 5375.187 MB
Total progress: 100%|██████████| 8/8 [00:03<00:00,  3.33it/s]
Total progress: 100%|██████████| 8/8 [00:03<00:00,  2.30it/s]
Prediction took 8.56 seconds

Examples

View more examples

Run time and cost

This model costs approximately $0.031 to run on Replicate, or 32 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 A100 (80GB) GPU hardware. Predictions typically complete within 23 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This model doesn't have a readme.