HunyuanVideo15Model<T>

Models & Types · AiDotNet.Diffusion.Video

HunyuanVideo 1.5 efficient video generation model for consumer GPUs.

Inheritance: DiffusionModelBase<T> → LatentDiffusionModelBase<T> → VideoDiffusionModelBase<T> → HunyuanVideo15Model<T>

For Beginners

HunyuanVideo 1.5 is a large (8.3B parameter) video generator that can run on consumer GPUs. It combines image and video generation in one model, understands complex text prompts via a multimodal language model encoder, and produces high-quality clips at up to 720p.

How It Works

References:

  • Paper: “HunyuanVideo 1.5: A Systematic Framework for Large Video Generation Model” (Tencent, 2025)

HunyuanVideo 1.5 is an efficient 8.3B parameter model achieving state-of-the-art visual quality while running on consumer GPUs. It combines a unified image-video generation architecture with a Causal 3D VAE and dual-stream DiT blocks. The model uses MLLM text encoding for precise prompt understanding and supports both T2V and I2V.

Technical specifications:

  • Architecture: Dual-Stream DiT + Causal 3D VAE + MLLM Text Encoder
  • Latent channels: 16
  • Default: 49 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, 49, 90, 160 });
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 HunyuanVideo15Model<float>(options: options))
    .Build(trainX, trainY);
var video = result.Predict(noise);

Constructors

Constructor Summary
HunyuanVideo15Model(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 HunyuanVideo15Model 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()