Freigeben über


Azure OpenAI (Preview)

Easily integrate Azure OpenAI's cutting-edge artificial intelligence capabilities into your workflows

This connector is available in the following products and regions:

Service Class Regions
Logic Apps Standard All Logic Apps regions except the following:
     -   Azure Government regions
     -   Azure China regions
     -   US Department of Defense (DoD)
Power Automate Premium All Power Automate regions except the following:
     -   US Government (GCC)
     -   US Government (GCC High)
     -   China Cloud operated by 21Vianet
     -   US Department of Defense (DoD)
Power Apps Premium All Power Apps regions except the following:
     -   US Government (GCC)
     -   US Government (GCC High)
     -   China Cloud operated by 21Vianet
     -   US Department of Defense (DoD)
Contact
Name Microsoft
URL https://support.microsoft.com
Connector Metadata
Publisher Abby Hartman
Website https://azure.microsoft.com/en-us/products/cognitive-services/openai-service
Privacy policy https://learn.microsoft.com/en-us/legal/cognitive-services/openai/data-privacy
Categories AI;Business Intelligence

Creating a connection

The connector supports the following authentication types:

Default Parameters for creating connection. All regions Not shareable

Default

Applicable: All regions

Parameters for creating connection.

This is not shareable connection. If the power app is shared with another user, another user will be prompted to create new connection explicitly.

Name Type Description Required
Azure OpenAI resource name string The name of the Azure OpenAI resource that hosts the AI model True
Azure OpenAI API key securestring The API key to access the Azure OpenAI resource that hosts the AI model True
Azure Cognitive Search endpoint URL string The URL of the Azure Cognitive Search endpoint indexing your data
Azure Cognitive Search API key securestring The API key to access the Azure Cognitive Search endpoint indexing your data

Throttling Limits

Name Calls Renewal Period
API calls per connection 1000 60 seconds

Actions

Creates a completion for the chat message

Creates a completion for the chat message

Get a vector representation of a given input that can be easily consumed by machine learning models and algorithms
Using extensions to create a completion for chat messages

Using extensions to create a completion for chat messages

Creates a completion for the chat message

Creates a completion for the chat message

Parameters

Name Key Required Type Description
Deployment ID of the deployed model
deployment-id True string

Deployment ID of the deployed model

API version
api-version True string

API version

temperature
temperature number

What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both.

top_p
top_p number

An 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. We generally recommend altering this or temperature but not both.

stream
stream boolean

If set, partial message deltas will be sent, like in ChatGPT. Tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a data: [DONE] message.

stop
stop array of string

Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence.

max_tokens
max_tokens integer

The maximum number of tokens allowed for the generated answer. By default, the number of tokens the model can return will be (4096 - prompt tokens).

presence_penalty
presence_penalty number

Number 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.

frequency_penalty
frequency_penalty number

Number 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.

logit_bias
logit_bias object

Modify the likelihood of specified tokens appearing in the completion. Accepts a json object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.

user
user string

A unique identifier representing your end-user, which can help Azure OpenAI to monitor and detect abuse.

role
role True string

The role of the messages author.

content
content True string

An array of content parts with a defined type, each can be of type text.

type
type True string

A representation of configuration data for a single Azure OpenAI chat extension. This will be used by a chat completions request that should use Azure OpenAI chat extensions to augment the response behavior. The use of this configuration is compatible only with Azure OpenAI.

top_n_documents
top_n_documents integer

The configured top number of documents to feature for the configured query.

in_scope
in_scope boolean

Whether queries should be restricted to use of indexed data.

strictness
strictness integer

The configured strictness of the search relevance filtering. The higher of strictness, the higher of the precision but lower recall of the answer.

role_information
role_information string

Give the model instructions about how it should behave and any context it should reference when generating a response. You can describe the assistant's personality and tell it how to format responses. There's a 100 token limit for it, and it counts against the overall token limit.

index_name
index_name True string

The name of the index to use as available in the referenced Azure Search resource.

title_field
title_field string

The name of the index field to use as a title.

url_field
url_field string

The name of the index field to use as a URL.

filepath_field
filepath_field string

The name of the index field to use as a filepath.

content_fields
content_fields array of string

The names of index fields that should be treated as content.

content_fields_separator
content_fields_separator string

The separator pattern that content fields should use.

vector_fields
vector_fields array of string

The names of fields that represent vector data.

query_type
query_type string

The type of Azure Search retrieval query that should be executed when using it as an Azure OpenAI chat extension.

semantic_configuration
semantic_configuration string

The additional semantic configuration for the query.

filter
filter string

Search filter.

type
type string

Represents the available sources Azure OpenAI On Your Data can use to configure vectorization of data for use with vector search.

deployment_name
deployment_name True string

Specifies the name of the model deployment to use for vectorization. This model deployment must be in the same Azure OpenAI resource, but On Your Data will use this model deployment via an internal call rather than a public one, which enables vector search even in private networks.

enabled
enabled boolean
enabled
enabled boolean
n
n integer

How many chat completion choices to generate for each input message.

seed
seed integer

If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result.Determinism is not guaranteed, and you should refer to the system_fingerprint response parameter to monitor changes in the backend.

logprobs
logprobs boolean

Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the content of message. This option is currently not available on the gpt-4-vision-preview model.

top_logprobs
top_logprobs integer

An integer between 0 and 5 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to true if this parameter is used.

type
type string

Setting to json_object enables JSON mode. This guarantees that the message the model generates is valid JSON.

Returns

Get a vector representation of a given input that can be easily consumed by machine learning models and algorithms

Parameters

Name Key Required Type Description
Deployment ID of the deployed model
deployment-id True string

Deployment ID of the deployed model

API version
api-version True string

API version

input
input True string

Input text to get embeddings for, encoded as a string. Input string must not exceed 2048 tokens in length

user
user string

A unique identifier representing your end-user, which can help Azure OpenAI to monitor and detect abuse.

input_type
input_type string

The input type of embedding search to use.

Returns

Using extensions to create a completion for chat messages

Using extensions to create a completion for chat messages

Parameters

Name Key Required Type Description
Deployment ID of the deployed model
deployment-id True string

Deployment ID of the deployed model

Confirm Deployment ID of the deployed model
deploymentId True string

Confirm Deployment ID of the deployed model

API version
api-version True string

API version

index
index integer

The index of the message in the conversation.

role
role True string

The role of the author of this message.

recipient
recipient string

The recipient of the message in the format of .. Present if and only if the recipient is tool.

content
content True string

The contents of the message

end_turn
end_turn boolean

Whether the message ends the turn.

type
type True string

The data source type.

parameters
parameters object

The parameters to be used for the data source in runtime.

temperature
temperature number

What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both.

top_p
top_p number

An 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. We generally recommend altering this or temperature but not both.

stream
stream boolean

If set, partial message deltas will be sent, like in ChatGPT. Tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a data: [DONE] message.

stop
stop array of string

Array minimum size of 1 and maximum of 4

max_tokens
max_tokens integer

The maximum number of tokens allowed for the generated answer. By default, the number of tokens the model can return will be (4096 - prompt tokens).

presence_penalty
presence_penalty number

Number 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.

frequency_penalty
frequency_penalty number

Number 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.

logit_bias
logit_bias object

Modify the likelihood of specified tokens appearing in the completion. Accepts a json object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.

user
user string

A unique identifier representing your end-user, which can help Azure OpenAI to monitor and detect abuse.

Returns

The response of the extensions chat completions.

Definitions

Message

A chat message.

Name Path Type Description
index
index integer

The index of the message in the conversation.

role
role string

The role of the author of this message.

recipient
recipient string

The recipient of the message in the format of .. Present if and only if the recipient is tool.

content
content string

The contents of the message

end_turn
end_turn boolean

Whether the message ends the turn.

ExtensionsChatCompletionsResponse

The response of the extensions chat completions.

Name Path Type Description
id
id string
object
object string
created
created integer
model
model string
prompt_filter_results
prompt_filter_results promptFilterResults

Content filtering results for zero or more prompts in the request. In a streaming request, results for different prompts may arrive at different times or in different orders.

prompt_tokens
usage.prompt_tokens integer
completion_tokens
usage.completion_tokens integer
total_tokens
usage.total_tokens integer
choices
choices array of ExtensionsChatCompletionChoice

ExtensionsChatCompletionChoice

Name Path Type Description
index
index integer
finish_reason
finish_reason string
content_filter_results
content_filter_results contentFilterResults

Information about the content filtering category (hate, sexual, violence, self_harm), if it has been detected, as well as the severity level (very_low, low, medium, high-scale that determines the intensity and risk level of harmful content) and if it has been filtered or not.

messages
messages array of Message

The list of messages returned by the service.

contentFilterResult

Name Path Type Description
severity
severity string
filtered
filtered boolean

contentFilterResults

Information about the content filtering category (hate, sexual, violence, self_harm), if it has been detected, as well as the severity level (very_low, low, medium, high-scale that determines the intensity and risk level of harmful content) and if it has been filtered or not.

Name Path Type Description
sexual
sexual contentFilterResult
violence
violence contentFilterResult
hate
hate contentFilterResult
self_harm
self_harm contentFilterResult
error
error errorBase

promptFilterResult

Content filtering results for a single prompt in the request.

Name Path Type Description
prompt_index
prompt_index integer
content_filter_results
content_filter_results contentFilterResults

Information about the content filtering category (hate, sexual, violence, self_harm), if it has been detected, as well as the severity level (very_low, low, medium, high-scale that determines the intensity and risk level of harmful content) and if it has been filtered or not.

promptFilterResults

Content filtering results for zero or more prompts in the request. In a streaming request, results for different prompts may arrive at different times or in different orders.

Name Path Type Description
Items
promptFilterResult

Content filtering results for a single prompt in the request.

errorBase

Name Path Type Description
code
code string
message
message string

errorBase_2024Feb15Preview

Name Path Type Description
code
code string
message
message string

contentFilterSeverityResult_2024Feb15Preview

Name Path Type Description
filtered
filtered boolean
severity
severity string

contentFilterDetectedResult_2024Feb15Preview

Name Path Type Description
filtered
filtered boolean
detected
detected boolean

contentFilterDetectedWithCitationResult_2024Feb15Preview

Name Path Type Description
filtered
filtered boolean
detected
detected boolean
URL
citation.URL string
license
citation.license string

contentFilterIdResult_2024Feb15Preview

Name Path Type Description
id
id string
filtered
filtered boolean

contentFilterPromptResults_2024Feb15Preview

Information about the content filtering category (hate, sexual, violence, self_harm), if it has been detected, as well as the severity level (very_low, low, medium, high-scale that determines the intensity and risk level of harmful content) and if it has been filtered or not. Information about jailbreak content and profanity, if it has been detected, and if it has been filtered or not. And information about customer block list, if it has been filtered and its id.

Name Path Type Description
sexual
sexual contentFilterSeverityResult_2024Feb15Preview
violence
violence contentFilterSeverityResult_2024Feb15Preview
hate
hate contentFilterSeverityResult_2024Feb15Preview
self_harm
self_harm contentFilterSeverityResult_2024Feb15Preview
profanity
profanity contentFilterDetectedResult_2024Feb15Preview
custom_blocklists
custom_blocklists array of contentFilterIdResult_2024Feb15Preview
error
error errorBase_2024Feb15Preview
jailbreak
jailbreak contentFilterDetectedResult_2024Feb15Preview

contentFilterChoiceResults_2024Feb15Preview

Information about the content filtering category (hate, sexual, violence, self_harm), if it has been detected, as well as the severity level (very_low, low, medium, high-scale that determines the intensity and risk level of harmful content) and if it has been filtered or not. Information about third party text and profanity, if it has been detected, and if it has been filtered or not. And information about customer block list, if it has been filtered and its id.

Name Path Type Description
sexual
sexual contentFilterSeverityResult_2024Feb15Preview
violence
violence contentFilterSeverityResult_2024Feb15Preview
hate
hate contentFilterSeverityResult_2024Feb15Preview
self_harm
self_harm contentFilterSeverityResult_2024Feb15Preview
profanity
profanity contentFilterDetectedResult_2024Feb15Preview
custom_blocklists
custom_blocklists array of contentFilterIdResult_2024Feb15Preview
error
error errorBase_2024Feb15Preview
protected_material_text
protected_material_text contentFilterDetectedResult_2024Feb15Preview
protected_material_code
protected_material_code contentFilterDetectedWithCitationResult_2024Feb15Preview

promptFilterResult_2024Feb15Preview

Content filtering results for a single prompt in the request.

Name Path Type Description
prompt_index
prompt_index integer
content_filter_results
content_filter_results contentFilterPromptResults_2024Feb15Preview

Information about the content filtering category (hate, sexual, violence, self_harm), if it has been detected, as well as the severity level (very_low, low, medium, high-scale that determines the intensity and risk level of harmful content) and if it has been filtered or not. Information about jailbreak content and profanity, if it has been detected, and if it has been filtered or not. And information about customer block list, if it has been filtered and its id.

promptFilterResults_2024Feb15Preview

Content filtering results for zero or more prompts in the request. In a streaming request, results for different prompts may arrive at different times or in different orders.

Name Path Type Description
Items
promptFilterResult_2024Feb15Preview

Content filtering results for a single prompt in the request.

azureChatExtensionsMessageContext_2024Feb15Preview

A representation of the additional context information available when Azure OpenAI chat extensions are involved in the generation of a corresponding chat completions response. This context information is only populated when using an Azure OpenAI request configured to use a matching extension.

Name Path Type Description
citations
citations array of citation_2024Feb15Preview

The data source retrieval result, used to generate the assistant message in the response.

intent
intent string

The detected intent from the chat history, used to pass to the next turn to carry over the context.

citation_2024Feb15Preview

citation information for a chat completions response message.

Name Path Type Description
content
content string

The content of the citation.

title
title string

The title of the citation.

url
url string

The URL of the citation.

filepath
filepath string

The file path of the citation.

chunk_id
chunk_id string

The chunk ID of the citation.

createChatCompletionResponse_2024Feb15Preview

Name Path Type Description
id
id string

A unique identifier for the chat completion.

object
object chatCompletionResponseObject_2024Feb15Preview

The object type.

created
created integer

The Unix timestamp (in seconds) of when the chat completion was created.

model
model string

The model used for the chat completion.

usage
usage completionUsage_2024Feb15Preview

Usage statistics for the completion request.

system_fingerprint
system_fingerprint string

Can be used in conjunction with the seed request parameter to understand when backend changes have been made that might impact determinism.

prompt_filter_results
prompt_filter_results promptFilterResults_2024Feb15Preview

Content filtering results for zero or more prompts in the request. In a streaming request, results for different prompts may arrive at different times or in different orders.

choices
choices array of object
index
choices.index integer
finish_reason
choices.finish_reason string
message
choices.message chatCompletionResponseMessage_2024Feb15Preview

A chat completion message generated by the model.

content_filter_results
choices.content_filter_results contentFilterChoiceResults_2024Feb15Preview

Information about the content filtering category (hate, sexual, violence, self_harm), if it has been detected, as well as the severity level (very_low, low, medium, high-scale that determines the intensity and risk level of harmful content) and if it has been filtered or not. Information about third party text and profanity, if it has been detected, and if it has been filtered or not. And information about customer block list, if it has been filtered and its id.

logprobs
choices.logprobs chatCompletionChoiceLogProbs_2024Feb15Preview

Log probability information for the choice.

chatCompletionChoiceLogProbs_2024Feb15Preview

Log probability information for the choice.

Name Path Type Description
content
content array of chatCompletionTokenLogprob_2024Feb15Preview

A list of message content tokens with log probability information.

chatCompletionTokenLogprob_2024Feb15Preview

Name Path Type Description
token
token string

The token.

logprob
logprob number

The log probability of this token.

bytes
bytes array of integer

A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token.

top_logprobs
top_logprobs array of object

List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned.

token
top_logprobs.token string

The token.

logprob
top_logprobs.logprob number

The log probability of this token.

bytes
top_logprobs.bytes array of integer

A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token.

chatCompletionResponseMessage_2024Feb15Preview

A chat completion message generated by the model.

Name Path Type Description
role
role chatCompletionResponseMessageRole_2024Feb15Preview

The role of the author of the response message.

content
content string

The contents of the message.

context
context azureChatExtensionsMessageContext_2024Feb15Preview

A representation of the additional context information available when Azure OpenAI chat extensions are involved in the generation of a corresponding chat completions response. This context information is only populated when using an Azure OpenAI request configured to use a matching extension.

chatCompletionResponseMessageRole_2024Feb15Preview

The role of the author of the response message.

The role of the author of the response message.

chatCompletionResponseObject_2024Feb15Preview

The object type.

The object type.

completionUsage_2024Feb15Preview

Usage statistics for the completion request.

Name Path Type Description
prompt_tokens
prompt_tokens integer

Number of tokens in the prompt.

completion_tokens
completion_tokens integer

Number of tokens in the generated completion.

total_tokens
total_tokens integer

Total number of tokens used in the request (prompt + completion).

getSingleEmbeddingsResponse_2024Feb15Preview

Name Path Type Description
object
object string
model
model string
data
data array of getEmbeddingsResponseDataItem
prompt_tokens
usage.prompt_tokens integer
total_tokens
usage.total_tokens integer

getEmbeddingsResponseDataItem

Name Path Type Description
index
index integer
object
object string
embedding
embedding array of number