Invoke-HuggingFaceInferenceApi.md

---

external help file: powershai-help.xml schema: 2.0.0 powershai: true

Invoke-HuggingFaceInferenceApi

SYNOPSIS

Invoca a API de Inferência Hugging Face https://huggingface.co/docs/hub/en/api

SYNTAX

Invoke-HuggingFaceInferenceApi [[-model] <Object>] [[-params] <Object>] [-Public] [-OpenaiChatCompletion] [[-StreamCallback] <Object>] 
[<CommonParameters>]

PARAMETERS

-model

Parameter Set: (All)
Type: Object
Aliases: 
Accepted Values: 
Required: false
Position: 1
Default Value: 
Accept pipeline input: false
Accept wildcard characters: false

-params

Parameter Set: (All)
Type: Object
Aliases: 
Accepted Values: 
Required: false
Position: 2
Default Value: 
Accept pipeline input: false
Accept wildcard characters: false

-Public

Parameter Set: (All)
Type: SwitchParameter
Aliases: 
Accepted Values: 
Required: false
Position: named
Default Value: False
Accept pipeline input: false
Accept wildcard characters: false

-OpenaiChatCompletion

Forçar usar endpoint de chat completion Params deverá ser tratado como o mesmo params da Api da Openai (Veja o cmdle Get-OpenaiChat). Mais info: https://huggingface.co/blog/tgi-messages-api So funciona com modelos que possuem um chat template!

Parameter Set: (All)
Type: SwitchParameter
Aliases: 
Accepted Values: 
Required: false
Position: named
Default Value: False
Accept pipeline input: false
Accept wildcard characters: false

-StreamCallback

Stream Callback para ser usado no caso de streamS!

Parameter Set: (All)
Type: Object
Aliases: 
Accepted Values: 
Required: false
Position: 3
Default Value: 
Accept pipeline input: false
Accept wildcard characters: false