SkyworkR1V<T>

Models & Types · AiDotNet.VisionLanguage.Reasoning

Skywork R1V: cross-modal transfer of reasoning LLMs to vision.

Inheritance: NeuralNetworkBase<T> → VisionLanguageModelBase<T> → SkyworkR1V<T>

Implements: IReasoningVLM<T>

For Beginners

Skywork R1V is a vision-language model that transfers reasoning capabilities from text LLMs to multimodal understanding. Default values follow the original paper settings.

How It Works

Skywork R1V (2025) pioneers cross-modal transfer of reasoning capabilities from text-only LLMs to vision-language models. It transfers chain-of-thought reasoning patterns learned by text reasoning models to the visual domain, enabling structured multi-step reasoning about images without requiring vision-specific reasoning data from scratch.

References:

  • Paper: “Skywork R1V: Pioneering Multimodal Reasoning with Chain-of-Thought” (2025)

Example

// Create a Skywork R1V model for multimodal chain-of-thought reasoning
// with cross-modal transfer of reasoning capabilities from LLMs to vision
var architecture = new NeuralNetworkArchitecture<double>(
    inputType: InputType.TwoDimensional,
    taskType: NeuralNetworkTaskType.ImageClassification,
    inputHeight: 224, inputWidth: 224, inputDepth: 3, outputSize: 512);
            
// ONNX inference mode with pre-trained model
var model = new SkyworkR1V<double>(architecture, "skyworkr1v.onnx");
            
// Training mode with native layers
var trainModel = new SkyworkR1V<double>(architecture, new SkyworkR1VOptions());

Constructors

Constructor Summary
SkyworkR1V(NeuralNetworkArchitecture<T> architecture, SkyworkR1VOptions options, IGradientBasedOptimizer<T, Tensor<T>, Tensor<T>> optimizer)
SkyworkR1V(NeuralNetworkArchitecture<T> architecture, string modelPath, SkyworkR1VOptions options)

Properties

Property Summary
int DecoderEmbeddingDim { get; }
int EmbeddingDimension { get; }
int MaxGenerationLength { get; }
string LanguageModelName { get; }
string ReasoningApproach { get; }

Methods

Method Summary
ModelMetadata<T> GetModelMetadata()
Tensor<T> Chat(Tensor<T> image, IEnumerable<ValueTuple<string, string>> conversationHistory, string userMessage)
Tensor<T> EncodeImage(Tensor<T> image)
Tensor<T> GenerateFromImage(Tensor<T> image, string prompt) Generates text from image using Skywork R1V’s cross-modal reasoning transfer.
Tensor<T> ReasonWithChainOfThought(Tensor<T> image, string question) Generates multi-step reasoning using Skywork R1V’s cross-modal transfer approach.