LumaRay3Model<T>

Models & Types · AiDotNet.Diffusion.Video

Luma Ray 3 with Hi-Fi Diffusion for 4K HDR video.

Inheritance: DiffusionModelBase<T> → LatentDiffusionModelBase<T> → VideoDiffusionModelBase<T> → LumaRay3Model<T>

For Beginners

Luma Ray 3 uses Hi-Fi Diffusion to generate videos at up to 4K HDR quality. It produces the highest resolution output among video generation models with exceptional visual clarity.

How It Works

References:

  • Reference: Luma AI Ray 3 (2025)

Ray 3 introduces Hi-Fi Diffusion technology that produces production-ready high-fidelity 4K HDR footage. The model packs significantly more detail in the same resolution compared to Ray 2, producing crisp outputs with state-of-the-art realism, physics, and character consistency.

Technical specifications:

  • Architecture: Hi-Fi Diffusion + 4K HDR + DiT
  • Latent channels: 16
  • Default: 120 frames at 24 FPS
  • Supports I2V: Yes | T2V: Yes | V2V: No

Example

var options = new DiffusionModelOptions<float> { LatentChannels = 16, DefaultInferenceSteps = 30 };
var noise = Tensor<float>.CreateRandom(new[] { 1, 16, 120, 270, 480 });
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 LumaRay3Model<float>(options: options))
    .Build(trainX, trainY);
var video = result.Predict(noise);

Constructors

Constructor Summary
LumaRay3Model(NeuralNetworkArchitecture<T> architecture, DiffusionModelOptions<T> options, INoiseScheduler<T> scheduler, DiTNoisePredictor<T> predictor, TemporalVAE<T> temporalVAE, IConditioningModule<T> conditioner, int defaultNumFrames, int defaultFPS) Initializes a new instance of LumaRay3Model with full customization support.

Properties

Property Summary
IConditioningModule<T> Conditioner { get; }
INoisePredictor<T> NoisePredictor { get; }
IVAEModel<T> TemporalVAE { get; }
IVAEModel<T> VAE { get; }
bool SupportsImageToVideo { get; }
bool SupportsTextToVideo { get; }
bool SupportsVideoToVideo { get; }
int LatentChannels { get; }

Methods

Method Summary
ModelMetadata<T> GetModelMetadata()