aaronaftab / mirage-ghibli

Ghiblify any image, 10x cheaper/faster than GPT 4o

Warm

Public
102.9K runs
H100
GitHub

Run with an API

Playground API Examples README Versions

Prediction

aaronaftab/mirage-ghibli:166efd15

Model

aaronaftab/mirage-ghibli:166efd15

5xy8qd4wt1rm80cnxwcv3gz5dr

Status

Succeeded

Source

Web

Hardware

H100

Total duration

12.2s

Created

2 days ago

Input

image
model: dev
prompt: GHBLI anime style photo
go_fast
lora_scale: 1
megapixels: 1
num_outputs: 1
aspect_ratio: 1:1
output_format: webp
guidance_scale: 10
output_quality: 80
prompt_strength: 0.78
extra_lora_scale: 1
num_inference_steps: 28

{
  "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
  "model": "dev",
  "prompt": "GHBLI anime style photo",
  "go_fast": true,
  "lora_scale": 1,
  "megapixels": "1",
  "num_outputs": 1,
  "aspect_ratio": "1:1",
  "output_format": "webp",
  "guidance_scale": 10,
  "output_quality": 80,
  "prompt_strength": 0.78,
  "extra_lora_scale": 1,
  "num_inference_steps": 28
}

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 aaronaftab/mirage-ghibli using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "aaronaftab/mirage-ghibli:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
  {
    input: {
      image: "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      model: "dev",
      prompt: "GHBLI anime style photo",
      go_fast: true,
      lora_scale: 1,
      megapixels: "1",
      num_outputs: 1,
      aspect_ratio: "1:1",
      output_format: "webp",
      guidance_scale: 10,
      output_quality: 80,
      prompt_strength: 0.78,
      extra_lora_scale: 1,
      num_inference_steps: 28
    }
  }
);
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 aaronaftab/mirage-ghibli using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "aaronaftab/mirage-ghibli:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
    input={
        "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
        "model": "dev",
        "prompt": "GHBLI anime style photo",
        "go_fast": True,
        "lora_scale": 1,
        "megapixels": "1",
        "num_outputs": 1,
        "aspect_ratio": "1:1",
        "output_format": "webp",
        "guidance_scale": 10,
        "output_quality": 80,
        "prompt_strength": 0.78,
        "extra_lora_scale": 1,
        "num_inference_steps": 28
    }
)
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 aaronaftab/mirage-ghibli 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": "166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
    "input": {
      "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      "model": "dev",
      "prompt": "GHBLI anime style photo",
      "go_fast": true,
      "lora_scale": 1,
      "megapixels": "1",
      "num_outputs": 1,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 10,
      "output_quality": 80,
      "prompt_strength": 0.78,
      "extra_lora_scale": 1,
      "num_inference_steps": 28
    }
  }' \
  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/aaronaftab/mirage-ghibli@sha256:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f \
  -i 'image="https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg"' \
  -i 'model="dev"' \
  -i 'prompt="GHBLI anime style photo"' \
  -i 'go_fast=true' \
  -i 'lora_scale=1' \
  -i 'megapixels="1"' \
  -i 'num_outputs=1' \
  -i 'aspect_ratio="1:1"' \
  -i 'output_format="webp"' \
  -i 'guidance_scale=10' \
  -i 'output_quality=80' \
  -i 'prompt_strength=0.78' \
  -i 'extra_lora_scale=1' \
  -i 'num_inference_steps=28'

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/aaronaftab/mirage-ghibli@sha256:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      "model": "dev",
      "prompt": "GHBLI anime style photo",
      "go_fast": true,
      "lora_scale": 1,
      "megapixels": "1",
      "num_outputs": 1,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 10,
      "output_quality": 80,
      "prompt_strength": 0.78,
      "extra_lora_scale": 1,
      "num_inference_steps": 28
    }
  }' \
  http://localhost:5000/predictions

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

Output

{
  "completed_at": "2025-04-01T02:06:40.229479Z",
  "created_at": "2025-04-01T02:06:28.048000Z",
  "data_removed": false,
  "error": null,
  "id": "5xy8qd4wt1rm80cnxwcv3gz5dr",
  "input": {
    "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
    "model": "dev",
    "prompt": "GHBLI anime style photo",
    "go_fast": true,
    "lora_scale": 1,
    "megapixels": "1",
    "num_outputs": 1,
    "aspect_ratio": "1:1",
    "output_format": "webp",
    "guidance_scale": 10,
    "output_quality": 80,
    "prompt_strength": 0.78,
    "extra_lora_scale": 1,
    "num_inference_steps": 28
  },
  "logs": "Img2img and inpainting not supported with fast fp8 inference; will run in bf16\nLoaded LoRAs in 0.61s\nUsing seed: 56976\nPrompt: GHBLI anime style photo\nInput image size: 768x1024\n[!] Resizing input image from 768x1024 to 768x1024\n[!] img2img mode\n  0%|          | 0/22 [00:00<?, ?it/s]\n  5%|▍         | 1/22 [00:00<00:03,  6.06it/s]\n  9%|▉         | 2/22 [00:00<00:03,  5.34it/s]\n 14%|█▎        | 3/22 [00:00<00:03,  5.15it/s]\n 18%|█▊        | 4/22 [00:00<00:03,  5.06it/s]\n 23%|██▎       | 5/22 [00:00<00:03,  5.02it/s]\n 27%|██▋       | 6/22 [00:01<00:03,  4.99it/s]\n 32%|███▏      | 7/22 [00:01<00:03,  4.97it/s]\n 36%|███▋      | 8/22 [00:01<00:02,  4.96it/s]\n 41%|████      | 9/22 [00:01<00:02,  4.95it/s]\n 45%|████▌     | 10/22 [00:01<00:02,  4.95it/s]\n 50%|█████     | 11/22 [00:02<00:02,  4.95it/s]\n 55%|█████▍    | 12/22 [00:02<00:02,  4.94it/s]\n 59%|█████▉    | 13/22 [00:02<00:01,  4.94it/s]\n 64%|██████▎   | 14/22 [00:02<00:01,  4.94it/s]\n 68%|██████▊   | 15/22 [00:03<00:01,  4.94it/s]\n 73%|███████▎  | 16/22 [00:03<00:01,  4.94it/s]\n 77%|███████▋  | 17/22 [00:03<00:01,  4.94it/s]\n 82%|████████▏ | 18/22 [00:03<00:00,  4.94it/s]\n 86%|████████▋ | 19/22 [00:03<00:00,  4.94it/s]\n 91%|█████████ | 20/22 [00:04<00:00,  4.94it/s]\n 95%|█████████▌| 21/22 [00:04<00:00,  4.94it/s]\n100%|██████████| 22/22 [00:04<00:00,  4.94it/s]\n100%|██████████| 22/22 [00:04<00:00,  4.98it/s]\nTotal safe images: 1 out of 1",
  "metrics": {
    "predict_time": 5.36541672,
    "total_time": 12.181479
  },
  "output": [
    "https://replicate.delivery/xezq/rsHvdh8bjbKQMh8BZdpBrp2cMnYe3UwdD4Jx0imbUdTY4HPKA/out-0.webp"
  ],
  "started_at": "2025-04-01T02:06:34.864062Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-rpq3ogm4stote4wf2gcgoyxp3m34yzo7xoucuaqdc3p5cyjwieda",
    "get": "https://api.replicate.com/v1/predictions/5xy8qd4wt1rm80cnxwcv3gz5dr",
    "cancel": "https://api.replicate.com/v1/predictions/5xy8qd4wt1rm80cnxwcv3gz5dr/cancel"
  },
  "version": "166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f"
}

Generated in

5.4 seconds

Tweak it ShareReport

Prediction

aaronaftab/mirage-ghibli:166efd15

Model

aaronaftab/mirage-ghibli:166efd15

gdcgke2ahnrme0cnxwdajzp1zg

Status

Succeeded

Source

Web

Hardware

H100

Total duration

6.9s

Created

2 days ago by @aaronaftab

Input

image
model: dev
prompt: GHBLI anime style photo
go_fast
lora_scale: 1
megapixels: 1
num_outputs: 1
aspect_ratio: 1:1
output_format: webp
guidance_scale: 10
output_quality: 80
prompt_strength: 0.8
extra_lora_scale: 1
num_inference_steps: 28

{
  "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
  "model": "dev",
  "prompt": "GHBLI anime style photo",
  "go_fast": true,
  "lora_scale": 1,
  "megapixels": "1",
  "num_outputs": 1,
  "aspect_ratio": "1:1",
  "output_format": "webp",
  "guidance_scale": 10,
  "output_quality": 80,
  "prompt_strength": 0.8,
  "extra_lora_scale": 1,
  "num_inference_steps": 28
}

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 aaronaftab/mirage-ghibli using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "aaronaftab/mirage-ghibli:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
  {
    input: {
      image: "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      model: "dev",
      prompt: "GHBLI anime style photo",
      go_fast: true,
      lora_scale: 1,
      megapixels: "1",
      num_outputs: 1,
      aspect_ratio: "1:1",
      output_format: "webp",
      guidance_scale: 10,
      output_quality: 80,
      prompt_strength: 0.8,
      extra_lora_scale: 1,
      num_inference_steps: 28
    }
  }
);
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 aaronaftab/mirage-ghibli using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "aaronaftab/mirage-ghibli:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
    input={
        "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
        "model": "dev",
        "prompt": "GHBLI anime style photo",
        "go_fast": True,
        "lora_scale": 1,
        "megapixels": "1",
        "num_outputs": 1,
        "aspect_ratio": "1:1",
        "output_format": "webp",
        "guidance_scale": 10,
        "output_quality": 80,
        "prompt_strength": 0.8,
        "extra_lora_scale": 1,
        "num_inference_steps": 28
    }
)
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 aaronaftab/mirage-ghibli 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": "166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
    "input": {
      "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      "model": "dev",
      "prompt": "GHBLI anime style photo",
      "go_fast": true,
      "lora_scale": 1,
      "megapixels": "1",
      "num_outputs": 1,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 10,
      "output_quality": 80,
      "prompt_strength": 0.8,
      "extra_lora_scale": 1,
      "num_inference_steps": 28
    }
  }' \
  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/aaronaftab/mirage-ghibli@sha256:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f \
  -i 'image="https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg"' \
  -i 'model="dev"' \
  -i 'prompt="GHBLI anime style photo"' \
  -i 'go_fast=true' \
  -i 'lora_scale=1' \
  -i 'megapixels="1"' \
  -i 'num_outputs=1' \
  -i 'aspect_ratio="1:1"' \
  -i 'output_format="webp"' \
  -i 'guidance_scale=10' \
  -i 'output_quality=80' \
  -i 'prompt_strength=0.8' \
  -i 'extra_lora_scale=1' \
  -i 'num_inference_steps=28'

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/aaronaftab/mirage-ghibli@sha256:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      "model": "dev",
      "prompt": "GHBLI anime style photo",
      "go_fast": true,
      "lora_scale": 1,
      "megapixels": "1",
      "num_outputs": 1,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 10,
      "output_quality": 80,
      "prompt_strength": 0.8,
      "extra_lora_scale": 1,
      "num_inference_steps": 28
    }
  }' \
  http://localhost:5000/predictions

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

Output

{
  "completed_at": "2025-04-01T02:07:19.425346Z",
  "created_at": "2025-04-01T02:07:12.525000Z",
  "data_removed": false,
  "error": null,
  "id": "gdcgke2ahnrme0cnxwdajzp1zg",
  "input": {
    "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
    "model": "dev",
    "prompt": "GHBLI anime style photo",
    "go_fast": true,
    "lora_scale": 1,
    "megapixels": "1",
    "num_outputs": 1,
    "aspect_ratio": "1:1",
    "output_format": "webp",
    "guidance_scale": 10,
    "output_quality": 80,
    "prompt_strength": 0.8,
    "extra_lora_scale": 1,
    "num_inference_steps": 28
  },
  "logs": "Img2img and inpainting not supported with fast fp8 inference; will run in bf16\nWeights already loaded\nLoaded LoRAs in 0.03s\nUsing seed: 34352\nPrompt: GHBLI anime style photo\nInput image size: 768x1024\n[!] Resizing input image from 768x1024 to 768x1024\n[!] img2img mode\n  0%|          | 0/23 [00:00<?, ?it/s]\n  4%|▍         | 1/23 [00:00<00:03,  6.05it/s]\n  9%|▊         | 2/23 [00:00<00:03,  5.33it/s]\n 13%|█▎        | 3/23 [00:00<00:03,  5.13it/s]\n 17%|█▋        | 4/23 [00:00<00:03,  5.04it/s]\n 22%|██▏       | 5/23 [00:00<00:03,  5.00it/s]\n 26%|██▌       | 6/23 [00:01<00:03,  4.97it/s]\n 30%|███       | 7/23 [00:01<00:03,  4.95it/s]\n 35%|███▍      | 8/23 [00:01<00:03,  4.94it/s]\n 39%|███▉      | 9/23 [00:01<00:02,  4.94it/s]\n 43%|████▎     | 10/23 [00:01<00:02,  4.93it/s]\n 48%|████▊     | 11/23 [00:02<00:02,  4.93it/s]\n 52%|█████▏    | 12/23 [00:02<00:02,  4.92it/s]\n 57%|█████▋    | 13/23 [00:02<00:02,  4.92it/s]\n 61%|██████    | 14/23 [00:02<00:01,  4.92it/s]\n 65%|██████▌   | 15/23 [00:03<00:01,  4.92it/s]\n 70%|██████▉   | 16/23 [00:03<00:01,  4.92it/s]\n 74%|███████▍  | 17/23 [00:03<00:01,  4.92it/s]\n 78%|███████▊  | 18/23 [00:03<00:01,  4.91it/s]\n 83%|████████▎ | 19/23 [00:03<00:00,  4.91it/s]\n 87%|████████▋ | 20/23 [00:04<00:00,  4.91it/s]\n 91%|█████████▏| 21/23 [00:04<00:00,  4.91it/s]\n 96%|█████████▌| 22/23 [00:04<00:00,  4.91it/s]\n100%|██████████| 23/23 [00:04<00:00,  4.91it/s]\n100%|██████████| 23/23 [00:04<00:00,  4.96it/s]\nTotal safe images: 1 out of 1",
  "metrics": {
    "predict_time": 5.00364167,
    "total_time": 6.900346
  },
  "output": [
    "https://replicate.delivery/xezq/EVLFXOA8IRJRMBq2qtFxEsNbq4c9E1n16BWWi4J1HOzV8jHF/out-0.webp"
  ],
  "started_at": "2025-04-01T02:07:14.421704Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-t5iobsjawmnm2lyun6wukfbroktvcpqn2jotxbnlbou6eiwwzn5a",
    "get": "https://api.replicate.com/v1/predictions/gdcgke2ahnrme0cnxwdajzp1zg",
    "cancel": "https://api.replicate.com/v1/predictions/gdcgke2ahnrme0cnxwdajzp1zg/cancel"
  },
  "version": "166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f"
}

Generated in

5.0 seconds

Tweak it ShareReport

Prediction

aaronaftab/mirage-ghibli:166efd15

Model

aaronaftab/mirage-ghibli:166efd15

pr3as3wek5rma0cnxwebs0x6r8

Status

Succeeded

Source

Web

Hardware

H100

Total duration

9.8s

Created

2 days ago by @aaronaftab

Input

image
model: dev
prompt: GHBLI anime style photo
go_fast
lora_scale: 1
megapixels: 1
num_outputs: 2
aspect_ratio: 1:1
output_format: webp
guidance_scale: 10
output_quality: 80
prompt_strength: 0.78
extra_lora_scale: 1
num_inference_steps: 28

{
  "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
  "model": "dev",
  "prompt": "GHBLI anime style photo",
  "go_fast": true,
  "lora_scale": 1,
  "megapixels": "1",
  "num_outputs": 2,
  "aspect_ratio": "1:1",
  "output_format": "webp",
  "guidance_scale": 10,
  "output_quality": 80,
  "prompt_strength": 0.78,
  "extra_lora_scale": 1,
  "num_inference_steps": 28
}

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 aaronaftab/mirage-ghibli using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "aaronaftab/mirage-ghibli:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
  {
    input: {
      image: "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      model: "dev",
      prompt: "GHBLI anime style photo",
      go_fast: true,
      lora_scale: 1,
      megapixels: "1",
      num_outputs: 2,
      aspect_ratio: "1:1",
      output_format: "webp",
      guidance_scale: 10,
      output_quality: 80,
      prompt_strength: 0.78,
      extra_lora_scale: 1,
      num_inference_steps: 28
    }
  }
);
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 aaronaftab/mirage-ghibli using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "aaronaftab/mirage-ghibli:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
    input={
        "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
        "model": "dev",
        "prompt": "GHBLI anime style photo",
        "go_fast": True,
        "lora_scale": 1,
        "megapixels": "1",
        "num_outputs": 2,
        "aspect_ratio": "1:1",
        "output_format": "webp",
        "guidance_scale": 10,
        "output_quality": 80,
        "prompt_strength": 0.78,
        "extra_lora_scale": 1,
        "num_inference_steps": 28
    }
)
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 aaronaftab/mirage-ghibli 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": "166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
    "input": {
      "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      "model": "dev",
      "prompt": "GHBLI anime style photo",
      "go_fast": true,
      "lora_scale": 1,
      "megapixels": "1",
      "num_outputs": 2,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 10,
      "output_quality": 80,
      "prompt_strength": 0.78,
      "extra_lora_scale": 1,
      "num_inference_steps": 28
    }
  }' \
  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/aaronaftab/mirage-ghibli@sha256:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f \
  -i 'image="https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg"' \
  -i 'model="dev"' \
  -i 'prompt="GHBLI anime style photo"' \
  -i 'go_fast=true' \
  -i 'lora_scale=1' \
  -i 'megapixels="1"' \
  -i 'num_outputs=2' \
  -i 'aspect_ratio="1:1"' \
  -i 'output_format="webp"' \
  -i 'guidance_scale=10' \
  -i 'output_quality=80' \
  -i 'prompt_strength=0.78' \
  -i 'extra_lora_scale=1' \
  -i 'num_inference_steps=28'

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/aaronaftab/mirage-ghibli@sha256:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      "model": "dev",
      "prompt": "GHBLI anime style photo",
      "go_fast": true,
      "lora_scale": 1,
      "megapixels": "1",
      "num_outputs": 2,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 10,
      "output_quality": 80,
      "prompt_strength": 0.78,
      "extra_lora_scale": 1,
      "num_inference_steps": 28
    }
  }' \
  http://localhost:5000/predictions

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

Output

{
  "completed_at": "2025-04-01T02:09:50.851640Z",
  "created_at": "2025-04-01T02:09:41.017000Z",
  "data_removed": false,
  "error": null,
  "id": "pr3as3wek5rma0cnxwebs0x6r8",
  "input": {
    "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
    "model": "dev",
    "prompt": "GHBLI anime style photo",
    "go_fast": true,
    "lora_scale": 1,
    "megapixels": "1",
    "num_outputs": 2,
    "aspect_ratio": "1:1",
    "output_format": "webp",
    "guidance_scale": 10,
    "output_quality": 80,
    "prompt_strength": 0.78,
    "extra_lora_scale": 1,
    "num_inference_steps": 28
  },
  "logs": "Img2img and inpainting not supported with fast fp8 inference; will run in bf16\nWeights already loaded\nLoaded LoRAs in 0.03s\nUsing seed: 27730\nPrompt: GHBLI anime style photo\nInput image size: 768x1024\n[!] Resizing input image from 768x1024 to 768x1024\n[!] img2img mode\n  0%|          | 0/22 [00:00<?, ?it/s]\n  5%|▍         | 1/22 [00:00<00:06,  3.28it/s]\n  9%|▉         | 2/22 [00:00<00:06,  2.86it/s]\n 14%|█▎        | 3/22 [00:01<00:06,  2.73it/s]\n 18%|█▊        | 4/22 [00:01<00:06,  2.68it/s]\n 23%|██▎       | 5/22 [00:01<00:06,  2.66it/s]\n 27%|██▋       | 6/22 [00:02<00:06,  2.64it/s]\n 32%|███▏      | 7/22 [00:02<00:05,  2.63it/s]\n 36%|███▋      | 8/22 [00:02<00:05,  2.62it/s]\n 41%|████      | 9/22 [00:03<00:04,  2.62it/s]\n 45%|████▌     | 10/22 [00:03<00:04,  2.62it/s]\n 50%|█████     | 11/22 [00:04<00:04,  2.62it/s]\n 55%|█████▍    | 12/22 [00:04<00:03,  2.61it/s]\n 59%|█████▉    | 13/22 [00:04<00:03,  2.61it/s]\n 64%|██████▎   | 14/22 [00:05<00:03,  2.61it/s]\n 68%|██████▊   | 15/22 [00:05<00:02,  2.61it/s]\n 73%|███████▎  | 16/22 [00:06<00:02,  2.61it/s]\n 77%|███████▋  | 17/22 [00:06<00:01,  2.61it/s]\n 82%|████████▏ | 18/22 [00:06<00:01,  2.60it/s]\n 86%|████████▋ | 19/22 [00:07<00:01,  2.61it/s]\n 91%|█████████ | 20/22 [00:07<00:00,  2.61it/s]\n 95%|█████████▌| 21/22 [00:07<00:00,  2.61it/s]\n100%|██████████| 22/22 [00:08<00:00,  2.61it/s]\n100%|██████████| 22/22 [00:08<00:00,  2.63it/s]\nTotal safe images: 2 out of 2",
  "metrics": {
    "predict_time": 9.827160750000001,
    "total_time": 9.83464
  },
  "output": [
    "https://replicate.delivery/xezq/Xe6XEYlmxHTQAKy7BIPDwfIzeXMT8Bi0YstCTYtg7X0dnf4RB/out-0.webp",
    "https://replicate.delivery/xezq/ZhBEh0BlBnKtKBPWR66VFUl6Xd8B1bSyyf7goG0oXqM35HPKA/out-1.webp"
  ],
  "started_at": "2025-04-01T02:09:41.024479Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-yvdbaipkkmukoqbetmy7eb3bnsju2yvcirr3aiv5i46cerq2q46q",
    "get": "https://api.replicate.com/v1/predictions/pr3as3wek5rma0cnxwebs0x6r8",
    "cancel": "https://api.replicate.com/v1/predictions/pr3as3wek5rma0cnxwebs0x6r8/cancel"
  },
  "version": "166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f"
}

Generated in

9.8 seconds

Tweak it ShareReport

Prediction

aaronaftab/mirage-ghibli:166efd15

Model

aaronaftab/mirage-ghibli:166efd15

gn15ezapfnrma0cnxwg8yprrbg

Status

Succeeded

Source

Web

Hardware

H100

Total duration

6.5s

Created

2 days ago

Input

image
model: dev
prompt: GHBLI anime style photo
go_fast
lora_scale: 1
megapixels: 1
num_outputs: 1
aspect_ratio: 1:1
output_format: webp
guidance_scale: 10
output_quality: 80
prompt_strength: 0.77
extra_lora_scale: 1
num_inference_steps: 38

{
  "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
  "model": "dev",
  "prompt": "GHBLI anime style photo",
  "go_fast": true,
  "lora_scale": 1,
  "megapixels": "1",
  "num_outputs": 1,
  "aspect_ratio": "1:1",
  "output_format": "webp",
  "guidance_scale": 10,
  "output_quality": 80,
  "prompt_strength": 0.77,
  "extra_lora_scale": 1,
  "num_inference_steps": 38
}

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 aaronaftab/mirage-ghibli using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "aaronaftab/mirage-ghibli:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
  {
    input: {
      image: "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      model: "dev",
      prompt: "GHBLI anime style photo",
      go_fast: true,
      lora_scale: 1,
      megapixels: "1",
      num_outputs: 1,
      aspect_ratio: "1:1",
      output_format: "webp",
      guidance_scale: 10,
      output_quality: 80,
      prompt_strength: 0.77,
      extra_lora_scale: 1,
      num_inference_steps: 38
    }
  }
);
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 aaronaftab/mirage-ghibli using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "aaronaftab/mirage-ghibli:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
    input={
        "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
        "model": "dev",
        "prompt": "GHBLI anime style photo",
        "go_fast": True,
        "lora_scale": 1,
        "megapixels": "1",
        "num_outputs": 1,
        "aspect_ratio": "1:1",
        "output_format": "webp",
        "guidance_scale": 10,
        "output_quality": 80,
        "prompt_strength": 0.77,
        "extra_lora_scale": 1,
        "num_inference_steps": 38
    }
)
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 aaronaftab/mirage-ghibli 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": "166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f",
    "input": {
      "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      "model": "dev",
      "prompt": "GHBLI anime style photo",
      "go_fast": true,
      "lora_scale": 1,
      "megapixels": "1",
      "num_outputs": 1,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 10,
      "output_quality": 80,
      "prompt_strength": 0.77,
      "extra_lora_scale": 1,
      "num_inference_steps": 38
    }
  }' \
  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/aaronaftab/mirage-ghibli@sha256:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f \
  -i 'image="https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg"' \
  -i 'model="dev"' \
  -i 'prompt="GHBLI anime style photo"' \
  -i 'go_fast=true' \
  -i 'lora_scale=1' \
  -i 'megapixels="1"' \
  -i 'num_outputs=1' \
  -i 'aspect_ratio="1:1"' \
  -i 'output_format="webp"' \
  -i 'guidance_scale=10' \
  -i 'output_quality=80' \
  -i 'prompt_strength=0.77' \
  -i 'extra_lora_scale=1' \
  -i 'num_inference_steps=38'

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/aaronaftab/mirage-ghibli@sha256:166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
      "model": "dev",
      "prompt": "GHBLI anime style photo",
      "go_fast": true,
      "lora_scale": 1,
      "megapixels": "1",
      "num_outputs": 1,
      "aspect_ratio": "1:1",
      "output_format": "webp",
      "guidance_scale": 10,
      "output_quality": 80,
      "prompt_strength": 0.77,
      "extra_lora_scale": 1,
      "num_inference_steps": 38
    }
  }' \
  http://localhost:5000/predictions

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

Output

{
  "completed_at": "2025-04-01T02:13:55.245619Z",
  "created_at": "2025-04-01T02:13:48.797000Z",
  "data_removed": false,
  "error": null,
  "id": "gn15ezapfnrma0cnxwg8yprrbg",
  "input": {
    "image": "https://replicate.delivery/pbxt/Mjo5VWBA0fm8oq3BefpbdtXeJitBJc6AobPnnMdFxxCKzT3P/linkedin.jpeg",
    "model": "dev",
    "prompt": "GHBLI anime style photo",
    "go_fast": true,
    "lora_scale": 1,
    "megapixels": "1",
    "num_outputs": 1,
    "aspect_ratio": "1:1",
    "output_format": "webp",
    "guidance_scale": 10,
    "output_quality": 80,
    "prompt_strength": 0.77,
    "extra_lora_scale": 1,
    "num_inference_steps": 38
  },
  "logs": "Img2img and inpainting not supported with fast fp8 inference; will run in bf16\nWeights already loaded\nLoaded LoRAs in 0.02s\nUsing seed: 4101\nPrompt: GHBLI anime style photo\nInput image size: 768x1024\n[!] Resizing input image from 768x1024 to 768x1024\n[!] img2img mode\n  0%|          | 0/30 [00:00<?, ?it/s]\n  3%|▎         | 1/30 [00:00<00:04,  6.03it/s]\n  7%|▋         | 2/30 [00:00<00:05,  5.31it/s]\n 10%|█         | 3/30 [00:00<00:05,  5.12it/s]\n 13%|█▎        | 4/30 [00:00<00:05,  5.03it/s]\n 17%|█▋        | 5/30 [00:00<00:05,  4.98it/s]\n 20%|██        | 6/30 [00:01<00:04,  4.95it/s]\n 23%|██▎       | 7/30 [00:01<00:04,  4.94it/s]\n 27%|██▋       | 8/30 [00:01<00:04,  4.93it/s]\n 30%|███       | 9/30 [00:01<00:04,  4.92it/s]\n 33%|███▎      | 10/30 [00:02<00:04,  4.91it/s]\n 37%|███▋      | 11/30 [00:02<00:03,  4.91it/s]\n 40%|████      | 12/30 [00:02<00:03,  4.91it/s]\n 43%|████▎     | 13/30 [00:02<00:03,  4.90it/s]\n 47%|████▋     | 14/30 [00:02<00:03,  4.90it/s]\n 50%|█████     | 15/30 [00:03<00:03,  4.90it/s]\n 53%|█████▎    | 16/30 [00:03<00:02,  4.90it/s]\n 57%|█████▋    | 17/30 [00:03<00:02,  4.90it/s]\n 60%|██████    | 18/30 [00:03<00:02,  4.90it/s]\n 63%|██████▎   | 19/30 [00:03<00:02,  4.90it/s]\n 67%|██████▋   | 20/30 [00:04<00:02,  4.90it/s]\n 70%|███████   | 21/30 [00:04<00:01,  4.90it/s]\n 73%|███████▎  | 22/30 [00:04<00:01,  4.90it/s]\n 77%|███████▋  | 23/30 [00:04<00:01,  4.90it/s]\n 80%|████████  | 24/30 [00:04<00:01,  4.90it/s]\n 83%|████████▎ | 25/30 [00:05<00:01,  4.90it/s]\n 87%|████████▋ | 26/30 [00:05<00:00,  4.90it/s]\n 90%|█████████ | 27/30 [00:05<00:00,  4.90it/s]\n 93%|█████████▎| 28/30 [00:05<00:00,  4.90it/s]\n 97%|█████████▋| 29/30 [00:05<00:00,  4.90it/s]\n100%|██████████| 30/30 [00:06<00:00,  4.90it/s]\n100%|██████████| 30/30 [00:06<00:00,  4.93it/s]\nTotal safe images: 1 out of 1",
  "metrics": {
    "predict_time": 6.440367618,
    "total_time": 6.448619
  },
  "output": [
    "https://replicate.delivery/xezq/gFJeG888ltyHEioGbyfeuMBI8euc2Hxfos00Gc6OpjzZ8ejHF/out-0.webp"
  ],
  "started_at": "2025-04-01T02:13:48.805251Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-utgont4f7o2jnefmhxaw4e7ayrpojinwacbg6hz3hqnzyq2vek6a",
    "get": "https://api.replicate.com/v1/predictions/gn15ezapfnrma0cnxwg8yprrbg",
    "cancel": "https://api.replicate.com/v1/predictions/gn15ezapfnrma0cnxwg8yprrbg/cancel"
  },
  "version": "166efd159b4138da932522bc5af40d39194033f587d9bdbab1e594119eae3e7f"
}

Generated in

6.5 seconds

Tweak it ShareReport

Img2img and inpainting not supported with fast fp8 inference; will run in bf16
Weights already loaded
Loaded LoRAs in 0.02s
Using seed: 4101
Prompt: GHBLI anime style photo
Input image size: 768x1024
[!] Resizing input image from 768x1024 to 768x1024
[!] img2img mode
  0%|          | 0/30 [00:00<?, ?it/s]
  3%|▎         | 1/30 [00:00<00:04,  6.03it/s]
  7%|▋         | 2/30 [00:00<00:05,  5.31it/s]
 10%|█         | 3/30 [00:00<00:05,  5.12it/s]
 13%|█▎        | 4/30 [00:00<00:05,  5.03it/s]
 17%|█▋        | 5/30 [00:00<00:05,  4.98it/s]
 20%|██        | 6/30 [00:01<00:04,  4.95it/s]
 23%|██▎       | 7/30 [00:01<00:04,  4.94it/s]
 27%|██▋       | 8/30 [00:01<00:04,  4.93it/s]
 30%|███       | 9/30 [00:01<00:04,  4.92it/s]
 33%|███▎      | 10/30 [00:02<00:04,  4.91it/s]
 37%|███▋      | 11/30 [00:02<00:03,  4.91it/s]
 40%|████      | 12/30 [00:02<00:03,  4.91it/s]
 43%|████▎     | 13/30 [00:02<00:03,  4.90it/s]
 47%|████▋     | 14/30 [00:02<00:03,  4.90it/s]
 50%|█████     | 15/30 [00:03<00:03,  4.90it/s]
 53%|█████▎    | 16/30 [00:03<00:02,  4.90it/s]
 57%|█████▋    | 17/30 [00:03<00:02,  4.90it/s]
 60%|██████    | 18/30 [00:03<00:02,  4.90it/s]
 63%|██████▎   | 19/30 [00:03<00:02,  4.90it/s]
 67%|██████▋   | 20/30 [00:04<00:02,  4.90it/s]
 70%|███████   | 21/30 [00:04<00:01,  4.90it/s]
 73%|███████▎  | 22/30 [00:04<00:01,  4.90it/s]
 77%|███████▋  | 23/30 [00:04<00:01,  4.90it/s]
 80%|████████  | 24/30 [00:04<00:01,  4.90it/s]
 83%|████████▎ | 25/30 [00:05<00:01,  4.90it/s]
 87%|████████▋ | 26/30 [00:05<00:00,  4.90it/s]
 90%|█████████ | 27/30 [00:05<00:00,  4.90it/s]
 93%|█████████▎| 28/30 [00:05<00:00,  4.90it/s]
 97%|█████████▋| 29/30 [00:05<00:00,  4.90it/s]
100%|██████████| 30/30 [00:06<00:00,  4.90it/s]
100%|██████████| 30/30 [00:06<00:00,  4.93it/s]
Total safe images: 1 out of 1

Want to make some of these yourself?

Run this model

aaronaftab / mirage-ghibli

Prediction

Input

Output

Prediction

Input

Output

Prediction

Input

Output

Prediction

Input

Output

Logs (5xy8qd4wt1rm80cnxwcv3gz5dr)

Logs (gdcgke2ahnrme0cnxwdajzp1zg)

Logs (pr3as3wek5rma0cnxwebs0x6r8)

Logs (gn15ezapfnrma0cnxwg8yprrbg)