Method LoadEncrypted

Namespace

LMKit.Model

Assembly

LM-Kit.NET.dll

LoadEncrypted(string, GgufEncryptionScheme, string, DeviceConfiguration, LoadingOptions, ModelLoadingProgressCallback)

Load a GGUF model from an LM-Kit encrypted container, decrypting tensor bytes on the fly from disk. The plaintext GGUF is never materialized in memory nor written back to disk: only the metadata block (a few MB) and one tensor's worth of bytes at a time are ever decrypted.

public static LM LoadEncrypted(string encryptedPath, GgufEncryptionScheme scheme, string password, LM.DeviceConfiguration deviceConfiguration = null, LM.LoadingOptions loadingOptions = null, LM.ModelLoadingProgressCallback loadingProgress = null)

Parameters

encryptedPath string

scheme GgufEncryptionScheme

password string

deviceConfiguration LM.DeviceConfiguration

loadingOptions LM.LoadingOptions

loadingProgress LM.ModelLoadingProgressCallback

Returns

LM

Remarks

Pass loadingOptions.LoadTensors = false to load only the metadata and vocabulary without decrypting any tensor bytes. This is the encrypted-model equivalent of the plaintext metadata-only path.

LoadEncrypted(string, string, DeviceConfiguration, LoadingOptions, ModelLoadingProgressCallback)

Load a GGUF model from an LM-Kit encrypted container, reading the encryption scheme from the container's own header. Equivalent to LoadEncrypted(string, GgufEncryptionScheme, string, DeviceConfiguration, LoadingOptions, ModelLoadingProgressCallback) with the recorded scheme.

public static LM LoadEncrypted(string encryptedPath, string password, LM.DeviceConfiguration deviceConfiguration = null, LM.LoadingOptions loadingOptions = null, LM.ModelLoadingProgressCallback loadingProgress = null)

Parameters

encryptedPath string

password string

deviceConfiguration LM.DeviceConfiguration

loadingOptions LM.LoadingOptions

loadingProgress LM.ModelLoadingProgressCallback

Returns

LM