ModelsLab/Stable Diffusion Trainer

sd-trainer

Stable Diffusion TrainerFree for Premium Users LLMs.txt

Efficiently train custom Stable Diffusion models with flexible batch sizes, gradient checkpointing, and memory-optimized attention requiring 12-24 GB VRAM for high-quality 512×512 to 1024×1024 image outputs.

API Playground API Documentation

API Endpoint URL

Base URL for all API requests to this endpoint.

https://modelslab.com/api/v6/trainer/train

API Authentication

Authentication requires a valid API key included in the request. Generate and manage your API keys from your developer dashboard. Include the key in the key parameter for all API requests.

Integration Examples

Production-ready code samples for API integration

{
  "instance_prompt": "thalman than",
  "images": [
    "https://pics.craiyon.com/2024-05-25/8sgy0rrpQvat_T1JLeAmEg.webp",
    "https://cdn.decohere.ai/cus_QmeRRs7CrpQL06-stablevideo/outputs/1728674089177/example_image.webp",
    "https://cdn.openart.ai/published/z1GNaRNOXOqkt1NmKac1/zdaB9HfQ_CAlV_1024.webp"
  ],
  "hf_token": null,
  "hf_username": null,
  "base_model_type": "sdxl",
  "seed": "298329",
  "training_steps": "1500",
  "resolution": "1024",
  "batch_size": "4",
  "mixed_precision": "fp16",
  "rank": "4",
  "alpha": "2",
  "learning_rate": "0.0001",
  "server_name": "NVIDIA GeForce RTX 3090 ($0.46/hr)",
  "trainer_id": "sd-trainer",
  "key": "YOUR_API_KEY"
}

SDKs

Official SDKs

Production-ready SDKs and client libraries for all major programming languages

JavaScript

Python

PHP

API Parameters

Technical specifications for API request parameters.

Field Name	Parameter	Technical Description
Instance Prompt	`instance_prompt`	Prompt for training the instance.
Images	`images`	Dataset for training, either as a path string or a list of paths (Required)
Hf Token	`hf_token`	Hugging Face token for accessing models.
Hf Username	`hf_username`	Hugging Face username for model storage
Base Model Type	`base_model_type`	Type of the base model. (Required)
Seed	`seed`	Random seed for reproducibility. (Required)
Training Steps	`training_steps`	No description available
Resolution	`resolution`	Resolution for the training images.
Batch Size	`batch_size`	Batch size for training.
Mixed Precision	`mixed_precision`	Precision mode for training allowed (fp16, bf16”)
Rank	`rank`	Rank for LoRA
Alpha	`alpha`	Alpha for LoRA.
Learning Rate	`learning_rate`	No description available
Server	`server_name`	No description available
Trainer Id	`trainer_id`	No description available