Method RunAsync

Namespace

LMKit.Extraction.Ocr

Assembly

LM-Kit.NET.dll

RunAsync(OcrParameters, CancellationToken)

Runs OCR on the specified page using the configured multimodal language model.

public override Task<OcrResult> RunAsync(OcrParameters ocrParameters, CancellationToken cancellationToken = default)

Parameters

ocrParameters OcrParameters

The OcrParameters containing the image data and attachment metadata for the page to be transcribed.

cancellationToken CancellationToken

A CancellationToken that can be used to cancel the operation.

Returns

Task<OcrResult>

A Task<TResult> that produces an OcrResult containing the page-level transcription returned by the model.

Remarks

This method implements the OcrEngine contract and is intended for consumers that only need the final OcrResult in a document-oriented workflow. For advanced scenarios that require access to model-level details, use Run(Attachment, CancellationToken) or RunAsync(Attachment, CancellationToken).

Exceptions

ArgumentNullException

Thrown if ocrParameters is null.

OperationCanceledException

Thrown if the operation is canceled via cancellationToken.

RunAsync(Attachment, CancellationToken)

Runs OCR on the specified attachment and returns a detailed result.

public Task<VlmOcr.VlmOcrResult> RunAsync(Attachment attachment, CancellationToken cancellationToken = default)

Parameters

attachment Attachment

The Attachment representing the image to be transcribed. The attachment may carry additional metadata (for example, page index or layout information) that can be used elsewhere in the extraction pipeline.

cancellationToken CancellationToken

A CancellationToken that can be used to cancel the operation.

Returns

Task<VlmOcr.VlmOcrResult>

A Task<TResult> that produces a VlmOcr.VlmOcrResult containing both the raw text-generation result and the constructed PageElement.

Exceptions

ArgumentNullException

Thrown if attachment is null.

OperationCanceledException

Thrown if the operation is canceled via cancellationToken.

RunAsync(ImageBuffer, CancellationToken)

Runs OCR on the specified image and returns a detailed result.

public Task<VlmOcr.VlmOcrResult> RunAsync(ImageBuffer image, CancellationToken cancellationToken = default)

Parameters

image ImageBuffer

The ImageBuffer containing the image to be transcribed.

cancellationToken CancellationToken

A CancellationToken that can be used to cancel the operation.

Returns

Task<VlmOcr.VlmOcrResult>

A Task<TResult> that produces a VlmOcr.VlmOcrResult containing both the raw text-generation result and the constructed PageElement.

Examples

var model = LM.LoadFromModelID("lmkit-tasks:4b-preview");
var ocr = new VlmOcr(model);

using var image = ImageBuffer.LoadAsRGB("invoice.png");
var result = await ocr.RunAsync(image, cancellationToken);

string text = result.PageElement.GetText();

Exceptions

ArgumentNullException

Thrown if image is null.

OperationCanceledException

Thrown if the operation is canceled via cancellationToken.