goodguy1963/sdxl-finetunes-img2img:08807c2a – Run with an API on Replicate

Version

You're looking at a specific version of this model. Jump to the model overview.

goodguy1963 /sdxl-finetunes-img2img:08807c2a

Playground API

Input

input_image

file

Input image for processing. If not provided, the default image will be used.

prompt

string

Shift + Return to add a new line

high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of fieldhigh quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field

High quality modifiers to add to prompt

Default: "high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field"

negative_prompt

string

Shift + Return to add a new line

ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposedugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed

Things you do not want to see in your image

Default: "ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed"

checkpoint_name

string

Checkpoint model to use

Default: "realismEngineSDXL_v30VAE.safetensors"

steps

integer

(minimum: 1, maximum: 100)

Number of inference steps

Default: 50

cfg

number

(minimum: 1, maximum: 20)

Guidance scale

Default: 7.5

denoise

number

(minimum: 0, maximum: 1)

Denoising strength

Default: 0.8

sampler_name

string

Sampling algorithm to use

Default: "euler"

scheduler

string

Scheduler for the sampler

Default: "karras"

lora1_strength_model

number

(minimum: 0, maximum: 2)

LoRA 1 (product marketing) model strength

Default: 1

lora1_strength_clip

number

(minimum: 0, maximum: 2)

LoRA 1 (product marketing) clip strength

Default: 1

lora2_strength_model

number

(minimum: 0, maximum: 2)

LoRA 2 (add-detail-xl) model strength

Default: 2

lora2_strength_clip

number

(minimum: 0, maximum: 2)

LoRA 2 (add-detail-xl) clip strength

Default: 1

lora3_strength_model

number

(minimum: 0, maximum: 2)

LoRA 3 (detailed_notrigger) model strength

Default: 1

lora3_strength_clip

number

(minimum: 0, maximum: 2)

LoRA 3 (detailed_notrigger) clip strength

Default: 1

ipadapter_weight

number

(minimum: 0, maximum: 1)

IP-Adapter weight

Default: 0.9

ipadapter_start_at

number

(minimum: 0, maximum: 1)

IP-Adapter start at step percentage

Default: 0

ipadapter_end_at

number

(minimum: 0, maximum: 1)

IP-Adapter end at step percentage

Default: 1

controlnet_strength

number

(minimum: 0, maximum: 2)

ControlNet strength

Default: 1

controlnet_start_percent

number

(minimum: 0, maximum: 1)

ControlNet start percentage

Default: 0

controlnet_end_percent

number

(minimum: 0, maximum: 1)

ControlNet end percentage

Default: 0.5

safety_filter

boolean

Enable Replicate safety filter to detect and filter NSFW content (can't be disabled in web UI)

Default: true

width

integer

(minimum: 512, maximum: 2048)

Width of the image

Default: 1024

height

integer

(minimum: 512, maximum: 2048)

Height of the image

Default: 1024

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: 95

seed

integer

Set a seed for reproducibility. Random by default.

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

npx create-replicate --model=goodguy1963/sdxl-finetunes-img2img

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 goodguy1963/sdxl-finetunes-img2img using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "goodguy1963/sdxl-finetunes-img2img:08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2",
  {
    input: {
      cfg: 7.5,
      steps: 50,
      width: 1024,
      height: 1024,
      prompt: "high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field",
      denoise: 0.8,
      scheduler: "karras",
      sampler_name: "euler",
      output_format: "webp",
      safety_filter: true,
      output_quality: 95,
      checkpoint_name: "realismEngineSDXL_v30VAE.safetensors",
      negative_prompt: "ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed",
      ipadapter_end_at: 1,
      ipadapter_weight: 0.9,
      ipadapter_start_at: 0,
      controlnet_strength: 1,
      lora1_strength_clip: 1,
      lora2_strength_clip: 1,
      lora3_strength_clip: 1,
      lora1_strength_model: 1,
      lora2_strength_model: 2,
      lora3_strength_model: 1,
      controlnet_end_percent: 0.5,
      controlnet_start_percent: 0
    }
  }
);

// 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 goodguy1963/sdxl-finetunes-img2img using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "goodguy1963/sdxl-finetunes-img2img:08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2",
    input={
        "cfg": 7.5,
        "steps": 50,
        "width": 1024,
        "height": 1024,
        "prompt": "high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field",
        "denoise": 0.8,
        "scheduler": "karras",
        "sampler_name": "euler",
        "output_format": "webp",
        "safety_filter": True,
        "output_quality": 95,
        "checkpoint_name": "realismEngineSDXL_v30VAE.safetensors",
        "negative_prompt": "ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed",
        "ipadapter_end_at": 1,
        "ipadapter_weight": 0.9,
        "ipadapter_start_at": 0,
        "controlnet_strength": 1,
        "lora1_strength_clip": 1,
        "lora2_strength_clip": 1,
        "lora3_strength_clip": 1,
        "lora1_strength_model": 1,
        "lora2_strength_model": 2,
        "lora3_strength_model": 1,
        "controlnet_end_percent": 0.5,
        "controlnet_start_percent": 0
    }
)
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 goodguy1963/sdxl-finetunes-img2img 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": "08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2",
    "input": {
      "cfg": 7.5,
      "steps": 50,
      "width": 1024,
      "height": 1024,
      "prompt": "high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field",
      "denoise": 0.8,
      "scheduler": "karras",
      "sampler_name": "euler",
      "output_format": "webp",
      "safety_filter": true,
      "output_quality": 95,
      "checkpoint_name": "realismEngineSDXL_v30VAE.safetensors",
      "negative_prompt": "ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed",
      "ipadapter_end_at": 1,
      "ipadapter_weight": 0.9,
      "ipadapter_start_at": 0,
      "controlnet_strength": 1,
      "lora1_strength_clip": 1,
      "lora2_strength_clip": 1,
      "lora3_strength_clip": 1,
      "lora1_strength_model": 1,
      "lora2_strength_model": 2,
      "lora3_strength_model": 1,
      "controlnet_end_percent": 0.5,
      "controlnet_start_percent": 0
    }
  }' \
  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/goodguy1963/sdxl-finetunes-img2img@sha256:08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2 \
  -i 'cfg=7.5' \
  -i 'steps=50' \
  -i 'width=1024' \
  -i 'height=1024' \
  -i 'prompt="high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field"' \
  -i 'denoise=0.8' \
  -i 'scheduler="karras"' \
  -i 'sampler_name="euler"' \
  -i 'output_format="webp"' \
  -i 'safety_filter=true' \
  -i 'output_quality=95' \
  -i 'checkpoint_name="realismEngineSDXL_v30VAE.safetensors"' \
  -i 'negative_prompt="ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed"' \
  -i 'ipadapter_end_at=1' \
  -i 'ipadapter_weight=0.9' \
  -i 'ipadapter_start_at=0' \
  -i 'controlnet_strength=1' \
  -i 'lora1_strength_clip=1' \
  -i 'lora2_strength_clip=1' \
  -i 'lora3_strength_clip=1' \
  -i 'lora1_strength_model=1' \
  -i 'lora2_strength_model=2' \
  -i 'lora3_strength_model=1' \
  -i 'controlnet_end_percent=0.5' \
  -i 'controlnet_start_percent=0'

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/goodguy1963/sdxl-finetunes-img2img@sha256:08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "cfg": 7.5,
      "steps": 50,
      "width": 1024,
      "height": 1024,
      "prompt": "high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field",
      "denoise": 0.8,
      "scheduler": "karras",
      "sampler_name": "euler",
      "output_format": "webp",
      "safety_filter": true,
      "output_quality": 95,
      "checkpoint_name": "realismEngineSDXL_v30VAE.safetensors",
      "negative_prompt": "ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed",
      "ipadapter_end_at": 1,
      "ipadapter_weight": 0.9,
      "ipadapter_start_at": 0,
      "controlnet_strength": 1,
      "lora1_strength_clip": 1,
      "lora2_strength_clip": 1,
      "lora3_strength_clip": 1,
      "lora1_strength_model": 1,
      "lora2_strength_model": 2,
      "lora3_strength_model": 1,
      "controlnet_end_percent": 0.5,
      "controlnet_start_percent": 0
    }
  }' \
  http://localhost:5000/predictions

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

Output

No output yet! Press "Submit" to start a prediction.