Skip to content

Azure OpenAI

Azure OpenAI is a research organization that aims to develop and direct artificial intelligence (AI) in ways that benefit humanity as a whole.

Categories: artificial-intelligence

Type: azureOpenAi/v1


Connections

Version: 1

Bearer Token

Properties

NameLabelTypeControl TypeDescriptionRequired
endpointEndpointSTRINGTEXTtrue
tokenTokenSTRINGTEXTtrue

Actions

Ask

Name: ask

Ask anything you want.

Properties

NameLabelTypeControl TypeDescriptionRequired
modelModelSTRINGTEXTDeployment name, written in string.true
messagesMessagesARRAY
Items [{STRING(role), STRING(content), [FILE_ENTRY](attachments)}]
ARRAY_BUILDERA list of messages comprising the conversation so far.true
responseResponseOBJECT
Properties {STRING(responseFormat), STRING(responseSchema)}
OBJECT_BUILDERThe response from the API.false
maxTokensMax TokensINTEGERINTEGERThe maximum number of tokens to generate in the chat completion.null
nNumber of Chat Completion ChoicesINTEGERINTEGERHow many chat completion choices to generate for each input message.null
temperatureTemperatureNUMBERNUMBERControls randomness: Higher values will make the output more random, while lower values like will make it more focused and deterministic.null
frequencyPenaltyFrequency PenaltyNUMBERNUMBERNumber between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model’s likelihood to repeat the same line verbatim.null
presencePenaltyPresence PenaltyNUMBERNUMBERNumber between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model’s likelihood to talk about new topics.null
logitBiasLogit BiasOBJECT
Properties {}
OBJECT_BUILDERModify the likelihood of specified tokens appearing in the completion.null
topPTop PNUMBERNUMBERAn alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.null
stopStopARRAY
Items [STRING]
ARRAY_BUILDERUp to 4 sequences where the API will stop generating further tokens.null
userUserSTRINGTEXTA unique identifier representing your end-user, which can help admins to monitor and detect abuse.false

JSON Example

{
"label" : "Ask",
"name" : "ask",
"parameters" : {
"model" : "",
"messages" : [ {
"role" : "",
"content" : "",
"attachments" : [ {
"extension" : "",
"mimeType" : "",
"name" : "",
"url" : ""
} ]
} ],
"response" : {
"responseFormat" : "",
"responseSchema" : ""
},
"maxTokens" : 1,
"n" : 1,
"temperature" : 0.0,
"frequencyPenalty" : 0.0,
"presencePenalty" : 0.0,
"logitBias" : { },
"topP" : 0.0,
"stop" : [ "" ],
"user" : ""
},
"type" : "azureOpenAi/v1/ask"
}

Create Image

Name: createImage

Create an image using text-to-image models

Properties

NameLabelTypeControl TypeDescriptionRequired
modelModelSTRING
Options dall-e-2, dall-e-3
SELECTThe model to use for image generation.true
imageMessagesMessagesARRAY
Items [{STRING(content), NUMBER(weight)}]
ARRAY_BUILDERA list of messages comprising the conversation so far.true
sizeSizeSTRING
Options DALL_E_2_256x256, DALL_E_2_512x512, _1024x1024, DALL_E_3_1792x1024, DALL_E_3_1024x1792
SELECTThe size of the generated images.true
nNumber of ResponsesINTEGERINTEGERThe number of images to generate. Must be between 1 and 10. For dall-e-3, only n=1 is supported..null
responseFormatResponse formatSTRING
Options URL, B64_JSON
SELECTThe format in which the generated images are returned.null
styleStyleSTRING
Options VIVID, NATURAL
SELECTThe style of the generated images. Must be one of vivid or natural. Vivid causes the model to lean towards generating hyper-real and dramatic images. Natural causes the model to produce more natural, less hyper-real looking images. This parameter is only supported for dall-e-3.null
userUserSTRINGTEXTA unique identifier representing your end-user, which can help admins to monitor and detect abuse.false

Output

Type: OBJECT

Properties

NameTypeControl Type
urlSTRINGTEXT
b64JsonSTRINGTEXT

JSON Example

{
"label" : "Create Image",
"name" : "createImage",
"parameters" : {
"model" : "",
"imageMessages" : [ {
"content" : "",
"weight" : 0.0
} ],
"size" : "",
"n" : 1,
"responseFormat" : "",
"style" : "",
"user" : ""
},
"type" : "azureOpenAi/v1/createImage"
}

Create Transcriptions

Name: createTranscription

Transcribes audio into the input language.

Properties

NameLabelTypeControl TypeDescriptionRequired
fileFileFILE_ENTRYFILE_ENTRYThe audio file object to transcribe, in one of these formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.true
modelModelSTRING
Options whisper
SELECTID of the model to use.true
languageLanguageSTRING
Options AF, AR, HY, AZ, BE, BS, BG, CA, ZH, HR, CS, DA, NL, EL, ET, EN, FI, FR, GL, DE, HE, HI, HU, IS, ID, IT, JA, KK, KN, KO, LT, LV, MA, MK, MR, MS, NE, NO, FA, PL, PT, RO, RU, SK, SL, SR, ES, SV, SW, TA, TL, TH, TR, UK, UR, VI, CY
SELECTThe language of the input audio.false
promptPromptSTRINGTEXTAn optional text to guide the model’s style or continue a previous audio segment. The prompt should match the audio language.false
responseFormatResponse FormatSTRING
Options JSON, SRT, TEXT, VERBOSE_JSON, VTT
SELECTThe format of the transcript outputtrue
temperatureTemperatureNUMBERNUMBERThe sampling temperature, between 0 and 1. Higher values like will make the output more random, while lower values will make it more focused and deterministic.false

JSON Example

{
"label" : "Create Transcriptions",
"name" : "createTranscription",
"parameters" : {
"file" : {
"extension" : "",
"mimeType" : "",
"name" : "",
"url" : ""
},
"model" : "",
"language" : "",
"prompt" : "",
"responseFormat" : "",
"temperature" : 0.0
},
"type" : "azureOpenAi/v1/createTranscription"
}