Functions/GenXdev.AI.LMStudio/Get-LMStudioTextEmbedding.ps1
|
function Get-LMStudioTextEmbedding { [CmdletBinding()] [Alias("embed-text", "Get-TextEmbedding")] param( ######################################################################## [Parameter( Mandatory = $true, Position = 0, HelpMessage = "Text to get embeddings for", ValueFromPipeline = $true )] [string[]] $Text, ######################################################################## [Parameter( Mandatory = $false, Position = 1, HelpMessage = "The LM-Studio model to use" )] [SupportsWildcards()] [string] $Model, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Identifier used for getting specific model from LM Studio" )] [string] $ModelLMSGetIdentifier, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Show the LM Studio window")] [switch] $ShowWindow, ######################################################################## [Alias("ttl")] [Parameter( Mandatory = $false, HelpMessage = "Set a TTL (in seconds) for models loaded via API requests")] [int] $TTLSeconds = -1, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "How much to offload to the GPU. If `"off`", GPU offloading is disabled. If `"max`", all layers are offloaded to GPU. If a number between 0 and 1, that fraction of layers will be offloaded to the GPU. -1 = LM Studio will decide how much to offload to the GPU. -2 = Auto " )] [int]$Gpu = -1, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Force stop LM Studio before initialization" )] [switch]$Force, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Api endpoint url, defaults to http://localhost:1234/v1/embeddings")] [string] $ApiEndpoint = $null, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "The API key to use for the request")] [string] $ApiKey = $null ######################################################################## ) begin { Microsoft.PowerShell.Utility\Write-Verbose "Starting text embedding process..." # initialize lm studio if using localhost if ([string]::IsNullOrWhiteSpace($ApiEndpoint) -or $ApiEndpoint.Contains("localhost")) { $initParams = GenXdev.Helpers\Copy-IdenticalParamValues ` -BoundParameters $PSBoundParameters ` -FunctionName 'GenXdev.AI\Initialize-LMStudioModel' ` -DefaultValues (Microsoft.PowerShell.Utility\Get-Variable -Scope Local -Name * ` -ErrorAction SilentlyContinue) if ($PSBoundParameters.ContainsKey("Force")) { $null = $PSBoundParameters.Remove("Force") $Force = $false } $modelInfo = GenXdev.AI\Initialize-LMStudioModel @initParams $Model = $modelInfo.identifier } # prepare api endpoint $apiUrl = "http://localhost:1234/v1/embeddings" if (-not [string]::IsNullOrWhiteSpace($ApiEndpoint)) { $apiUrl = $ApiEndpoint } $headers = @{ "Content-Type" = "application/json" } if (-not [string]::IsNullOrWhiteSpace($ApiKey)) { $headers."Authorization" = "Bearer $ApiKey" } } process { Microsoft.PowerShell.Utility\Write-Verbose "Getting embeddings for text with model: $Model" # prepare api payload $payload = @{ model = $Model input = $Text } # convert payload to json $json = $payload | Microsoft.PowerShell.Utility\ConvertTo-Json -Depth 60 -Compress $bytes = [System.Text.Encoding]::UTF8.GetBytes($json) Microsoft.PowerShell.Utility\Write-Verbose "Requesting embeddings from LM-Studio model '$Model'" # send request with long timeouts $response = Microsoft.PowerShell.Utility\Invoke-RestMethod -Uri $apiUrl ` -Method Post ` -Body $bytes ` -Headers $headers ` -OperationTimeoutSeconds (3600 * 24) ` -ConnectionTimeoutSeconds (3600 * 24) # Output embedding data foreach ($embedding in $response.data) { [PSCustomObject]@{ embedding = $embedding.embedding index = $embedding.index text = $Text[$embedding.index] } } } } |