Class LM.LoadingOptions

Namespace

LMKit.Model

Assembly

LM-Kit.NET.dll

Represents options for loading a model, allowing customization of the loading process.

public sealed class LM.LoadingOptions

Inheritance

object

LM.LoadingOptions

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.ReferenceEquals(object, object)

object.ToString()

Properties

EnableCpuWeightRepacking

Whether weights placed in host memory may be repacked into CPU-optimized layouts (a large matrix-multiplication speedup for CPU inference). Repacked layouts serve forward inference only: fine-tuning cannot read them, so a model intended for TRAINING on the CPU must be loaded with this set to false. Models fully offloaded to a GPU are unaffected either way. Defaults to true.

EnableSpeculativeDecodingDrafts

Gets or sets a value indicating whether the speculative decoding draft assets packaged with the model are loaded for this LM instance.

LoadTensors

Gets or sets a value indicating whether model weights should be loaded. Setting this property to false allows for quicker access to model properties like metadata and vocabulary without loading the weights.

MetadataOverrides

Gets or sets a collection of metadata overrides to apply during the model loading process.

Methods

Clone()

Creates a deep clone of the current LM.LoadingOptions instance.