Skip to main content

Train a Lora Model with Custom Images

Overview​

Using this endpoint you can train a Lora model with your own images. You can train a model on any object or person.

Request​

--request POST 'https://stablediffusionapi.com/api/v3/lora_fine_tune' \

Make a POST request to https://stablediffusionapi.com/api/v3/lora_fine_tune endpoint and pass the required parameters as a request body.

For now, you can only train a model on normal lora models and sdxl and get its style.

Body Attributes​

ParameterDescription
keyYour API Key used for request authorization
instance_promptText prompt with how you want to call your trained person/object
class_promptClassification of the trained person/object
base_model_typeThe type of lora base model you want to train on. It accepts normal or sdxl
negative_promptItems you don't want in the image
imagesPass accessible direct links to images, cropped to 512 x 512 px. A good number is about 7-8 images.
training_typeThe type of the object you are training on
lora_typeIt accepts lora or lycoris
max_train_stepsSet it at 2 times the number of images (Ni*2; maximum value 2000)
webhookGet a post call when training is complete

Training Types​

The table below lists all the possible values for the training_type parameter.

ValueDescription
menTrain on faces of men.
femaleTrain on faces of females.
coupleTrain on couples of male and female; in images array pass images of couples, instead of images of a single person.
nullTrain on object or anything.

Webhook Post JSON​

This is an example webhook post call in JSON format.

{
 "status": "success",
 "training_status": "deploying_gpu",
 "logs": "it will take upto 25 minutes",
 "model_id": "F5jvdzGnYi",
}

Training Status Values​

The table below describes all possible training statuses.

StatusDescription
deploying_gpuDeploying GPU
training_startedTraining started
training_successTraining completed successfully
trained_model_compressingCompressing the trained model
trained_model_uploadingUploading the trained model
trained_model_uploadedTrained model uploaded
deploying_modelDeploying the trained model
model_readyThe trained model is ready for use

Example​

Body​

Body
{
    "key":"",
    "instance_prompt": "photo of ambika0 man",
    "class_prompt": "photo of a man",
    "base_model_type": "sdxl",
    "negative_prompt":" lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry",
    "images": [
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/1.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/2.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/3.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/4.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/5.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/6.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/7.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/8.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/9.png"
  ],
    "seed": "0",
    "training_type": "men",
    "max_train_steps": "2",
    "lora_type":"lora",
    "webhook": null
}

Request​

var myHeaders = new Headers();
myHeaders.append("Content-Type", "application/json");

var raw = JSON.stringify({
    "key":"",
    "instance_prompt": "photo of ambika0 man",
    "class_prompt": "photo of a man",
    "base_model_type": "sdxl",
    "negative_prompt":" lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry",
    "images": [
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/1.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/2.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/3.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/4.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/5.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/6.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/7.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/8.png",
    "https://raw.githubusercontent.com/pnavitha/sampleImages/master/9.png"
  ],
    "seed": "0",
    "training_type": "men",
    "max_train_steps": "2",
    "lora_type":"lora",
    "webhook": null
});

var requestOptions = {
  method: 'POST',
  headers: myHeaders,
  body: raw,
  redirect: 'follow'
};

fetch("https://stablediffusionapi.com/api/v3/lora_fine_tune", requestOptions)
  .then(response => response.text())
  .then(result => console.log(result))
  .catch(error => console.log('error', error));

Response​

{
  "status": "success",
  "messege": "deploying_gpu",
  "data": "it will take upto 30 minutes.",
  "training_id": "F5jvdzGnYi"
}