Constructor DocumentSplitting

Namespace

LMKit.Extraction

Assembly

LM-Kit.NET.dll

DocumentSplitting(LM)

Initializes a new instance of the DocumentSplitting class with the specified vision language model.

public DocumentSplitting(LM model)

Parameters

model LM

A vision-capable LM model to use for boundary detection; cannot be null. The model must support both text generation (HasTextGeneration) and vision (HasVision).

Examples

// Load a vision-capable model (8B or larger recommended for document splitting)
LM model = LM.LoadFromModelID("qwen3-vl:8b");

// Create a DocumentSplitting instance
DocumentSplitting splitter = new DocumentSplitting(model);

// Optionally configure and run
splitter.Guidance = "The file contains invoices and contracts.";
DocumentSplittingResult result = splitter.Split(new Attachment("batch.pdf"));
Console.WriteLine($"Found {result.DocumentCount} document(s)");

Exceptions

ArgumentNullException

Thrown when the provided model is null.

InvalidModelException

Thrown when the model does not support text generation, does not have vision capabilities, or has been loaded without weights.