alexgenovese/train-sdxl-kohya:d23600f6 | Run with an API on Replicate

Version

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

alexgenovese /train-sdxl-kohya:d23600f6

Input

pretrained_model_name_or_path

string

Shift + Return to add a new line

stabilityai/stable-diffusion-xl-base-1.0stabilityai/stable-diffusion-xl-base-1.0

base model name or path

Default: "stabilityai/stable-diffusion-xl-base-1.0"

train_data_zip

*file

Upload image dataset in zip format using this naming convention: XX_token className.zip

network_weights

file

Pretrained LoRA weights

output_name

string

Shift + Return to add a new line

Model name

Default: "new_model_name"

save_model_as

string

model save extension | ckpt, pt, safetensors

Default: "safetensors"

resolution

string

Shift + Return to add a new line

image resolution must be 'size' or 'width,height'.

Default: "1024"

batch_size

integer

(minimum: 1)

batch size

Default: 1

max_train_epoches

integer

(minimum: 1)

max train epoches

Default: 10

save_every_n_epochs

integer

(minimum: 1)

save every n epochs

Default: 2

network_dim

integer

(minimum: 1)

network dimension

Default: 32

network_alpha

integer

(minimum: 1)

network alpha

Default: 32

train_unet_only

boolean

train U-Net only

Default: false

train_text_encoder_only

boolean

train Text Encoder only

Default: false

seed

integer

(minimum: 1)

reproducable seed

Default: 98796

noise_offset

number

(minimum: 0, maximum: 1)

noise offset

Default: 0

keep_tokens

integer

(minimum: 0)

keep heading N tokens when shuffling caption tokens

Default: 0

learning_rate

number

(minimum: 0)

Learning rate

Default: 0.00006

unet_lr

number

(minimum: 0)

UNet learning rate

Default: 0.00006

text_encoder_lr

number

(minimum: 0)

Text Encoder learning rate

Default: 0.000007

lr_scheduler

string

"linear", "cosine", "cosine_with_restarts", "polynomial", "constant", "constant_with_warmup

Default: "cosine"

lr_warmup_steps

integer

(minimum: 0)

warmup steps

Default: 0

lr_scheduler_num_cycles

integer

(minimum: 1)

cosine_with_restarts restart cycles

Default: 1

min_bucket_reso

integer

(minimum: 1)

arb min resolution

Default: 256

max_bucket_reso

integer

(minimum: 1)

arb max resolution

Default: 1024

persistent_data_loader_workers

boolean

makes workers persistent, further reduces/eliminates the lag in between epochs. however it may increase memory usage

Default: true

clip_skip

integer

(minimum: 0)

clip skip

Default: 2

optimizer_type

string

adaFactor","AdamW","AdamW8bit","Lion","SGDNesterov","SGDNesterov8bit","DAdaptation", "Lion", "Prodigy

Default: "Lion"

network_module

string

Network module

Default: "networks.lora"

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

npx create-replicate --model=alexgenovese/train-sdxl-kohya

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";
import fs from "node:fs";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run alexgenovese/train-sdxl-kohya using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "alexgenovese/train-sdxl-kohya:d23600f6a8dd645a59bbcdeb2ceac4ef5546ba7c931077b8fa3b95fd4cb25b49",
  {
    input: {
      seed: 98796,
      unet_lr: 0.00006,
      clip_skip: 2,
      batch_size: 1,
      resolution: "1024",
      keep_tokens: 0,
      network_dim: 32,
      output_name: "new_model_name",
      lr_scheduler: "cosine",
      noise_offset: 0,
      learning_rate: 0.00006,
      network_alpha: 32,
      save_model_as: "safetensors",
      network_module: "networks.lora",
      optimizer_type: "Lion",
      lr_warmup_steps: 0,
      max_bucket_reso: 1024,
      min_bucket_reso: 256,
      text_encoder_lr: 0.000007,
      train_unet_only: false,
      max_train_epoches: 10,
      save_every_n_epochs: 2,
      lr_scheduler_num_cycles: 1,
      train_text_encoder_only: false,
      pretrained_model_name_or_path: "stabilityai/stable-diffusion-xl-base-1.0",
      persistent_data_loader_workers: true
    }
  }
);

// To access the file URL:
console.log(output.url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", 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 alexgenovese/train-sdxl-kohya using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "alexgenovese/train-sdxl-kohya:d23600f6a8dd645a59bbcdeb2ceac4ef5546ba7c931077b8fa3b95fd4cb25b49",
    input={
        "seed": 98796,
        "unet_lr": 0.00006,
        "clip_skip": 2,
        "batch_size": 1,
        "resolution": "1024",
        "keep_tokens": 0,
        "network_dim": 32,
        "output_name": "new_model_name",
        "lr_scheduler": "cosine",
        "noise_offset": 0,
        "learning_rate": 0.00006,
        "network_alpha": 32,
        "save_model_as": "safetensors",
        "network_module": "networks.lora",
        "optimizer_type": "Lion",
        "lr_warmup_steps": 0,
        "max_bucket_reso": 1024,
        "min_bucket_reso": 256,
        "text_encoder_lr": 0.000007,
        "train_unet_only": False,
        "max_train_epoches": 10,
        "save_every_n_epochs": 2,
        "lr_scheduler_num_cycles": 1,
        "train_text_encoder_only": False,
        "pretrained_model_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
        "persistent_data_loader_workers": True
    }
)
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 alexgenovese/train-sdxl-kohya 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": "alexgenovese/train-sdxl-kohya:d23600f6a8dd645a59bbcdeb2ceac4ef5546ba7c931077b8fa3b95fd4cb25b49",
    "input": {
      "seed": 98796,
      "unet_lr": 0.00006,
      "clip_skip": 2,
      "batch_size": 1,
      "resolution": "1024",
      "keep_tokens": 0,
      "network_dim": 32,
      "output_name": "new_model_name",
      "lr_scheduler": "cosine",
      "noise_offset": 0,
      "learning_rate": 0.00006,
      "network_alpha": 32,
      "save_model_as": "safetensors",
      "network_module": "networks.lora",
      "optimizer_type": "Lion",
      "lr_warmup_steps": 0,
      "max_bucket_reso": 1024,
      "min_bucket_reso": 256,
      "text_encoder_lr": 0.000007,
      "train_unet_only": false,
      "max_train_epoches": 10,
      "save_every_n_epochs": 2,
      "lr_scheduler_num_cycles": 1,
      "train_text_encoder_only": false,
      "pretrained_model_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
      "persistent_data_loader_workers": true
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

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