ControlNetLiteModel<T>

Models & Types · AiDotNet.Diffusion.Control

Lightweight ControlNet model with reduced parameter count for faster inference.

Inheritance: DiffusionModelBase<T> → LatentDiffusionModelBase<T> → ControlNetLiteModel<T>

For Beginners

This is a smaller, faster version of ControlNet. It uses fewer parameters so it runs quicker and uses less memory, at the cost of slightly less precise control signal adherence.

How It Works

ControlNet Lite reduces the encoder to approximately 25% of the full ControlNet’s parameters while maintaining acceptable control quality. Suitable for real-time or resource-constrained applications.

Example

// Create a lightweight ControlNet for fast inference
var options = new DiffusionModelOptions<float>
{
    LatentChannels = 4,
    DefaultInferenceSteps = 15
};
            
// Generate with lightweight depth-guided control
var depthMap = Tensor<float>.CreateRandom(new[] { 1, 1, 512, 512 });
var trainX = Tensor<float>.CreateRandom(4, 8);
var trainY = Tensor<float>.CreateRandom(4, 2);
var result = new AiModelBuilder<float, Tensor<float>, Tensor<float>>()
    .ConfigureModel(new ControlNetLiteModel<float>(options: options, controlType: ControlType.Depth))
    .Build(trainX, trainY);
var prediction = result.Predict(depthMap);

Constructors

Constructor Summary
ControlNetLiteModel(NeuralNetworkArchitecture<T> architecture, DiffusionModelOptions<T> options, INoiseScheduler<T> scheduler, UNetNoisePredictor<T> baseUNet, StandardVAE<T> vae, IConditioningModule<T> conditioner, ControlType controlType, int? seed) Initializes a new ControlNet Lite model.

Properties

Property Summary
IConditioningModule<T> Conditioner { get; }
INoisePredictor<T> NoisePredictor { get; }
IVAEModel<T> VAE { get; }
int LatentChannels { get; }

Methods

Method Summary
ModelMetadata<T> GetModelMetadata()