GrokVision<T>

Models & Types · AiDotNet.VisionLanguage.Proprietary

Grok Vision: reference implementation of xAI’s real-time multimodal model.

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

Implements: IProprietaryVLM<T>

For Beginners

Grok Vision is a proprietary multimodal model from xAI with real-time data access and visual understanding. Default values follow the model’s recommended settings.

How It Works

Grok Vision is a reference implementation of xAI’s real-time multimodal model. Grok processes images alongside text with access to real-time data, featuring a large context window and multimodal understanding capabilities for visual question answering, image analysis, and document understanding.

References:

  • Grok Vision: real-time data processing with multimodal input (xAI, 2024-2025)

Example

// Create a Grok Vision model for real-time multimodal understanding
// with real-time data processing and multimodal input from xAI
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 GrokVision<double>(architecture, "grokvision.onnx");
            
// Training mode with native layers
var trainModel = new GrokVision<double>(architecture, new GrokVisionOptions());

Constructors

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

Properties

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

Methods

Method Summary
ModelMetadata<T> GetModelMetadata()
Tensor<T> Chat(Tensor<T> image, string prompt)
Tensor<T> EncodeImage(Tensor<T> image)
Tensor<T> GenerateFromImage(Tensor<T> image, string prompt) Generates text from an image using Grok’s real-time multimodal architecture.