> ## Documentation Index > Fetch the complete documentation index at: https://azure-foundry.mintlify.site/llms.txt > Use this file to discover all available pages before exploring further. # Post chatcompletions > Creates a chat completion. ## OpenAPI ````yaml api-reference/openai-v1.yaml post /chat/completions openapi: 3.0.0 info: title: Azure AI Foundry Models Service license: name: MIT url: https://github.com/openai/openai-openapi/blob/master/LICENSE version: v1 servers: - url: '{endpoint}/openai/v1' description: Azure AI Foundry Models APIs variables: endpoint: default: '' description: >- A supported Azure AI Foundry Models APIs endpoint, including protocol and hostname. For example: https://westus.api.cognitive.microsoft.com). security: - ApiKeyAuth: [] - ApiKeyAuth_: [] - OAuth2Auth: - https://cognitiveservices.azure.com/.default tags: - name: Chat - name: Embeddings - name: Evals - name: Files - name: Fine-tuning - name: Models - name: Responses - name: Vector Stores paths: /chat/completions: post: tags: - Chat description: Creates a chat completion. operationId: createChatCompletion parameters: - name: api-version in: query required: false description: >- The explicit Azure AI Foundry Models API version to use for this request. `v1` if not otherwise specified. schema: $ref: '#/components/schemas/AzureAIFoundryModelsApiVersion' default: v1 requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/AzureCreateChatCompletionRequest' responses: '200': description: The request has succeeded. headers: apim-request-id: required: false description: A request ID used for troubleshooting purposes. schema: type: string content: application/json: schema: $ref: '#/components/schemas/AzureCreateChatCompletionResponse' text/event-stream: schema: $ref: '#/components/schemas/AzureCreateChatCompletionStreamResponse' default: description: An unexpected error response. headers: apim-request-id: required: false description: A request ID used for troubleshooting purposes. schema: type: string content: application/json: schema: $ref: '#/components/schemas/AzureErrorResponse' components: schemas: AzureAIFoundryModelsApiVersion: type: string enum: - v1 - preview AzureCreateChatCompletionRequest: type: object required: - messages - model properties: metadata: type: object additionalProperties: type: string description: >- Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format, and querying for objects via API or the dashboard. Keys are strings with a maximum length of 64 characters. Values are strings with a maximum length of 512 characters. x-oaiTypeLabel: map temperature: type: number format: float nullable: true minimum: 0 maximum: 2 description: >- 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. default: 1 top_p: type: number format: float nullable: true minimum: 0 maximum: 1 description: >- 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. default: 1 user: type: string description: |- A unique identifier representing your end-user, which can help to monitor and detect abuse. top_logprobs: type: integer format: int32 minimum: 0 maximum: 20 description: >- An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. messages: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionRequestMessage' minItems: 1 description: >- A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, images, and audio. modalities: type: object allOf: - $ref: '#/components/schemas/ResponseModalities' nullable: true reasoning_effort: allOf: - $ref: '#/components/schemas/OpenAI.ReasoningEffort' nullable: true default: medium max_completion_tokens: type: integer format: int32 nullable: true description: |- An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens. frequency_penalty: type: number format: float nullable: true minimum: -2 maximum: 2 description: >- 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. default: 0 presence_penalty: type: number format: float nullable: true minimum: -2 maximum: 2 description: >- 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. default: 0 response_format: allOf: - $ref: '#/components/schemas/OpenAI.ResponseFormat' description: >- An object specifying the format that the model must output. Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured Outputs which ensures the model will match your supplied JSON schema. Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON. **Important:** when using JSON mode, you **must** also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if `finish_reason="length"`, which indicates the generation exceeded `max_tokens` or the conversation exceeded the max context length. audio: type: object properties: voice: allOf: - $ref: '#/components/schemas/OpenAI.VoiceIdsShared' description: >- The voice the model uses to respond. Supported voices are `alloy`, `ash`, `ballad`, `coral`, `echo`, `fable`, `nova`, `onyx`, `sage`, and `shimmer`. format: type: string enum: - wav - aac - mp3 - flac - opus - pcm16 description: >- Specifies the output audio format. Must be one of `wav`, `mp3`, `flac`, `opus`, or `pcm16`. required: - voice - format nullable: true description: >- Parameters for audio output. Required when audio output is requested with `modalities: ["audio"]`. store: type: boolean nullable: true description: >- Whether or not to store the output of this chat completion request for use in model distillation or evals products. default: false stream: type: boolean nullable: true description: >- If set to true, the model response data will be streamed to the client as it is generated using server-sent events. default: false stop: allOf: - $ref: '#/components/schemas/OpenAI.StopConfiguration' nullable: true default: null logit_bias: type: object additionalProperties: type: integer format: int32 nullable: true description: >- 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. x-oaiTypeLabel: map default: null logprobs: type: boolean nullable: true description: >- 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`. default: false max_tokens: type: integer format: int32 nullable: true description: >- The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API. This value is now deprecated in favor of `max_completion_tokens`, and is not compatible with o1 series models. deprecated: true 'n': type: integer format: int32 nullable: true minimum: 1 maximum: 128 description: >- How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep `n` as `1` to minimize costs. default: 1 prediction: type: object allOf: - $ref: '#/components/schemas/OpenAI.ChatOutputPrediction' nullable: true description: |- Configuration for a predicted output, which can greatly improve response times when large parts of the model response are known ahead of time. This is most common when you are regenerating a file with only minor changes to most of the content. seed: type: integer format: int64 nullable: true minimum: -9223372036854776000 maximum: 9223372036854776000 description: >- This feature is in Beta. 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. stream_options: type: object allOf: - $ref: '#/components/schemas/OpenAI.ChatCompletionStreamOptions' nullable: true default: null tools: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionTool' description: >- A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported. tool_choice: $ref: '#/components/schemas/OpenAI.ChatCompletionToolChoiceOption' parallel_tool_calls: allOf: - $ref: '#/components/schemas/OpenAI.ParallelToolCalls' default: true function_call: anyOf: - type: string enum: - none - auto - $ref: '#/components/schemas/OpenAI.ChatCompletionFunctionCallOption' description: >- Deprecated in favor of `tool_choice`. Controls which (if any) function is called by the model. `none` means the model will not call a function and instead generates a message. `auto` means the model can pick between generating a message or calling a function. Specifying a particular function via `{"name": "my_function"}` forces the model to call that function. `none` is the default when no functions are present. `auto` is the default if functions are present. deprecated: true functions: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionFunctions' minItems: 1 maxItems: 128 description: |- Deprecated in favor of `tools`. A list of functions the model may generate JSON inputs for. deprecated: true model: type: string description: >- The model deployment identifier to use for the chat completion request. data_sources: type: array items: $ref: '#/components/schemas/AzureChatDataSource' description: >- The data sources to use for the On Your Data feature, exclusive to Azure OpenAI. user_security_context: $ref: '#/components/schemas/AzureUserSecurityContext' description: >- The extended request model for chat completions against the Azure OpenAI service. This adds the ability to provide data sources for the On Your Data feature. AzureCreateChatCompletionResponse: type: object required: - id - created - model - object - choices properties: id: type: string description: A unique identifier for the chat completion. created: type: integer format: unixtime description: >- The Unix timestamp (in seconds) of when the chat completion was created. model: type: string description: The model used for the chat completion. system_fingerprint: type: string description: >- This fingerprint represents the backend configuration that the model runs with. Can be used in conjunction with the `seed` request parameter to understand when backend changes have been made that might impact determinism. object: type: string enum: - chat.completion description: The object type, which is always `chat.completion`. usage: $ref: '#/components/schemas/OpenAI.CompletionUsage' choices: type: array items: type: object properties: finish_reason: type: string enum: - stop - length - tool_calls - content_filter - function_call description: >- The reason the model stopped generating tokens. This will be `stop` if the model hit a natural stop point or a provided stop sequence, `length` if the maximum number of tokens specified in the request was reached, `content_filter` if content was omitted due to a flag from our content filters, `tool_calls` if the model called a tool, or `function_call` (deprecated) if the model called a function. index: type: integer format: int32 description: The index of the choice in the list of choices. logprobs: type: object properties: content: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionTokenLogprob' nullable: true description: >- A list of message content tokens with log probability information. readOnly: true refusal: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionTokenLogprob' nullable: true description: >- A list of message refusal tokens with log probability information. readOnly: true required: - content - refusal nullable: true description: Log probability information for the choice. content_filter_results: $ref: '#/components/schemas/AzureContentFilterResultForChoice' message: allOf: - $ref: '#/components/schemas/AzureChatCompletionResponseMessage' description: The chat completion response message. required: - finish_reason - index - logprobs - message prompt_filter_results: type: array items: type: object properties: prompt_index: type: integer format: int32 description: >- The index of the input prompt that this content filter result corresponds to. content_filter_results: allOf: - $ref: '#/components/schemas/AzureContentFilterResultForPrompt' description: >- The content filter results associated with the indexed input prompt. required: - prompt_index - content_filter_results description: >- The Responsible AI content filter annotations associated with prompt inputs into chat completions. description: >- The extended top-level chat completion response model for the Azure OpenAI service. This model adds Responsible AI content filter annotations for prompt input. AzureCreateChatCompletionStreamResponse: type: object required: - id - choices - created - model - object properties: id: type: string description: >- A unique identifier for the chat completion. Each chunk has the same ID. choices: type: array items: type: object properties: delta: $ref: '#/components/schemas/OpenAI.ChatCompletionStreamResponseDelta' logprobs: type: object properties: content: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionTokenLogprob' nullable: true description: >- A list of message content tokens with log probability information. readOnly: true refusal: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionTokenLogprob' nullable: true description: >- A list of message refusal tokens with log probability information. readOnly: true required: - content - refusal nullable: true description: Log probability information for the choice. finish_reason: type: string enum: - stop - length - tool_calls - content_filter - function_call nullable: true description: >- The reason the model stopped generating tokens. This will be `stop` if the model hit a natural stop point or a provided stop sequence, `length` if the maximum number of tokens specified in the request was reached, `content_filter` if content was omitted due to a flag from our content filters, `tool_calls` if the model called a tool, or `function_call` (deprecated) if the model called a function. index: type: integer format: int32 description: The index of the choice in the list of choices. required: - delta - finish_reason - index description: >- A list of chat completion choices. Can contain more than one elements if `n` is greater than 1. Can also be empty for the last chunk if you set `stream_options: {"include_usage": true}`. created: type: integer format: unixtime description: >- The Unix timestamp (in seconds) of when the chat completion was created. Each chunk has the same timestamp. model: type: string description: The model to generate the completion. system_fingerprint: type: string description: >- This fingerprint represents the backend configuration that the model runs with. Can be used in conjunction with the `seed` request parameter to understand when backend changes have been made that might impact determinism. object: type: string enum: - chat.completion.chunk description: The object type, which is always `chat.completion.chunk`. usage: type: object allOf: - $ref: '#/components/schemas/OpenAI.CompletionUsage' nullable: true description: >- An optional field that will only be present when you set `stream_options: {"include_usage": true}` in your request. When present, it contains a null value **except for the last chunk** which contains the token usage statistics for the entire request. **NOTE:** If the stream is interrupted or cancelled, you may not receive the final usage chunk which contains the total token usage for the request. delta: $ref: '#/components/schemas/AzureChatCompletionStreamResponseDelta' content_filter_results: $ref: '#/components/schemas/AzureContentFilterResultForChoice' AzureErrorResponse: type: object properties: error: type: object properties: code: type: string description: The distinct, machine-generated identifier for the error. message: type: string description: A human-readable message associated with the error. param: type: string description: >- If applicable, the request input parameter associated with the error type: type: string enum: - error description: The object type, always 'error.' inner_error: {} description: The error details. OpenAI.ChatCompletionRequestMessage: type: object required: - role properties: role: allOf: - $ref: '#/components/schemas/OpenAI.ChatCompletionRole' description: The role of the author of this message. content: anyOf: - type: string - type: array items: $ref: >- #/components/schemas/OpenAI.ChatCompletionRequestMessageContentPart nullable: true description: >- The content of the message. Valid content part types vary per message role. discriminator: propertyName: role mapping: system: '#/components/schemas/OpenAI.ChatCompletionRequestSystemMessage' developer: '#/components/schemas/OpenAI.ChatCompletionRequestDeveloperMessage' user: '#/components/schemas/OpenAI.ChatCompletionRequestUserMessage' assistant: '#/components/schemas/OpenAI.ChatCompletionRequestAssistantMessage' tool: '#/components/schemas/OpenAI.ChatCompletionRequestToolMessage' function: '#/components/schemas/OpenAI.ChatCompletionRequestFunctionMessage' ResponseModalities: type: array items: type: string enum: - text - audio description: >- Output types that you would like the model to generate. Most models are capable of generating text, which is the default: `["text"]` The `gpt-4o-audio-preview` model can also be used to [generate audio](/docs/guides/audio). To request that this model generate both text and audio responses, you can use: `["text", "audio"]` OpenAI.ReasoningEffort: type: string enum: - low - medium - high description: |- **o-series models only** Constrains effort on reasoning for [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently supported values are `low`, `medium`, and `high`. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response. OpenAI.ResponseFormat: type: object required: - type properties: type: type: string enum: - text - json_object - json_schema discriminator: propertyName: type mapping: text: '#/components/schemas/OpenAI.ResponseFormatText' json_object: '#/components/schemas/OpenAI.ResponseFormatJsonObject' json_schema: '#/components/schemas/OpenAI.ResponseFormatJsonSchema' OpenAI.VoiceIdsShared: anyOf: - type: string - type: string enum: - alloy - ash - ballad - coral - echo - fable - onyx - nova - sage - shimmer - verse OpenAI.StopConfiguration: anyOf: - type: string - type: array items: type: string description: |- Not supported with latest reasoning models `o3` and `o4-mini`. Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence. OpenAI.ChatOutputPrediction: type: object required: - type properties: type: $ref: '#/components/schemas/OpenAI.ChatOutputPredictionType' discriminator: propertyName: type mapping: content: '#/components/schemas/OpenAI.ChatOutputPredictionContent' description: Base representation of predicted output from a model. OpenAI.ChatCompletionStreamOptions: type: object properties: include_usage: type: boolean description: >- If set, an additional chunk will be streamed before the `data: [DONE]` message. The `usage` field on this chunk shows the token usage statistics for the entire request, and the `choices` field will always be an empty array. All other chunks will also include a `usage` field, but with a null value. **NOTE:** If the stream is interrupted, you may not receive the final usage chunk which contains the total token usage for the request. description: >- Options for streaming response. Only set this when you set `stream: true`. OpenAI.ChatCompletionTool: type: object required: - type - function properties: type: type: string enum: - function description: The type of the tool. Currently, only `function` is supported. function: $ref: '#/components/schemas/OpenAI.FunctionObject' OpenAI.ChatCompletionToolChoiceOption: anyOf: - type: string enum: - none - auto - required - $ref: '#/components/schemas/OpenAI.ChatCompletionNamedToolChoice' description: >- Controls which (if any) tool is called by the model. `none` means the model will not call any tool and instead generates a message. `auto` means the model can pick between generating a message or calling one or more tools. `required` means the model must call one or more tools. Specifying a particular tool via `{"type": "function", "function": {"name": "my_function"}}` forces the model to call that tool. `none` is the default when no tools are present. `auto` is the default if tools are present. OpenAI.ParallelToolCalls: type: boolean description: >- Whether to enable [parallel function calling](/docs/guides/function-calling#configuring-parallel-function-calling) during tool use. OpenAI.ChatCompletionFunctionCallOption: type: object required: - name properties: name: type: string description: The name of the function to call. description: >- Specifying a particular function via `{"name": "my_function"}` forces the model to call that function. OpenAI.ChatCompletionFunctions: type: object required: - name properties: description: type: string description: >- A description of what the function does, used by the model to choose when and how to call the function. name: type: string description: >- The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64. parameters: description: >- The parameters the functions accepts, described as a JSON Schema object. See the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format. Omitting `parameters` defines a function with an empty parameter list. deprecated: true AzureChatDataSource: type: object required: - type properties: type: allOf: - $ref: '#/components/schemas/AzureChatDataSourceType' description: The differentiating type identifier for the data source. discriminator: propertyName: type mapping: azure_search: '#/components/schemas/AzureSearchChatDataSource' azure_cosmos_db: '#/components/schemas/AzureCosmosDBChatDataSource' elasticsearch: '#/components/schemas/ElasticsearchChatDataSource' pinecone: '#/components/schemas/PineconeChatDataSource' mongo_db: '#/components/schemas/MongoDBChatDataSource' description: >- A representation of configuration data for a single Azure OpenAI chat data source. 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. AzureUserSecurityContext: type: object properties: application_name: type: string description: >- The name of the application. Sensitive personal information should not be included in this field. end_user_id: type: string description: >- This identifier is the Microsoft Entra ID (formerly Azure Active Directory) user object ID used to authenticate end-users within the generative AI application. Sensitive personal information should not be included in this field. end_user_tenant_id: type: string description: >- The Microsoft 365 tenant ID the end user belongs to. It's required when the generative AI application is multitenant. source_ip: type: string description: Captures the original client's IP address. description: >- User security context contains several parameters that describe the application itself, and the end user that interacts with the application. These fields assist your security operations teams to investigate and mitigate security incidents by providing a comprehensive approach to protecting your AI applications. [Learn more](https://aka.ms/TP4AI/Documentation/EndUserContext) about protecting AI applications using Microsoft Defender for Cloud. OpenAI.CompletionUsage: type: object required: - completion_tokens - prompt_tokens - total_tokens properties: completion_tokens: type: integer format: int32 description: Number of tokens in the generated completion. default: 0 prompt_tokens: type: integer format: int32 description: Number of tokens in the prompt. default: 0 total_tokens: type: integer format: int32 description: Total number of tokens used in the request (prompt + completion). default: 0 completion_tokens_details: type: object properties: accepted_prediction_tokens: type: integer format: int32 description: |- When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion. default: 0 audio_tokens: type: integer format: int32 description: Audio input tokens generated by the model. default: 0 reasoning_tokens: type: integer format: int32 description: Tokens generated by the model for reasoning. default: 0 rejected_prediction_tokens: type: integer format: int32 description: >- When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits. default: 0 description: Breakdown of tokens used in a completion. prompt_tokens_details: type: object properties: audio_tokens: type: integer format: int32 description: Audio input tokens present in the prompt. default: 0 cached_tokens: type: integer format: int32 description: Cached tokens present in the prompt. default: 0 description: Breakdown of tokens used in the prompt. description: Usage statistics for the completion request. OpenAI.ChatCompletionTokenLogprob: type: object required: - token - logprob - bytes - top_logprobs properties: token: type: string description: The token. logprob: type: number format: float description: >- The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value `-9999.0` is used to signify that the token is very unlikely. bytes: type: array items: type: integer format: int32 nullable: true description: >- 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: type: array items: type: object properties: token: type: string description: The token. logprob: type: number format: float description: >- The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value `-9999.0` is used to signify that the token is very unlikely. bytes: type: array items: type: integer format: int32 nullable: true description: >- 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. required: - token - logprob - bytes description: >- 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. AzureContentFilterResultForChoice: type: object properties: sexual: allOf: - $ref: '#/components/schemas/AzureContentFilterSeverityResult' description: >- A content filter category for language related to anatomical organs and genitals, romantic relationships, acts portrayed in erotic or affectionate terms, pregnancy, physical sexual acts, including those portrayed as an assault or a forced sexual violent act against one's will, prostitution, pornography, and abuse. hate: allOf: - $ref: '#/components/schemas/AzureContentFilterSeverityResult' description: >- A content filter category that can refer to any content that attacks or uses pejorative or discriminatory language with reference to a person or identity group based on certain differentiating attributes of these groups including but not limited to race, ethnicity, nationality, gender identity and expression, sexual orientation, religion, immigration status, ability status, personal appearance, and body size. violence: allOf: - $ref: '#/components/schemas/AzureContentFilterSeverityResult' description: >- A content filter category for language related to physical actions intended to hurt, injure, damage, or kill someone or something; describes weapons, guns and related entities, such as manufactures, associations, legislation, and so on. self_harm: allOf: - $ref: '#/components/schemas/AzureContentFilterSeverityResult' description: >- A content filter category that describes language related to physical actions intended to purposely hurt, injure, damage one's body or kill oneself. profanity: allOf: - $ref: '#/components/schemas/AzureContentFilterDetectionResult' description: >- A detection result that identifies whether crude, vulgar, or otherwise objection language is present in the content. custom_blocklists: allOf: - $ref: '#/components/schemas/AzureContentFilterBlocklistResult' description: >- A collection of binary filtering outcomes for configured custom blocklists. custom_topics: allOf: - $ref: '#/components/schemas/AzureContentFilterCustomTopicResult' description: >- A collection of binary filtering outcomes for configured custom topics. error: type: object properties: code: type: integer format: int32 description: A distinct, machine-readable code associated with the error. message: type: string description: A human-readable message associated with the error. required: - code - message description: >- If present, details about an error that prevented content filtering from completing its evaluation. protected_material_text: allOf: - $ref: '#/components/schemas/AzureContentFilterDetectionResult' description: >- A detection result that describes a match against text protected under copyright or other status. protected_material_code: type: object properties: filtered: type: boolean description: >- Whether the content detection resulted in a content filtering action. detected: type: boolean description: >- Whether the labeled content category was detected in the content. citation: type: object properties: license: type: string description: >- The name or identifier of the license associated with the detection. URL: type: string format: uri description: The URL associated with the license. description: >- If available, the citation details describing the associated license and its location. required: - filtered - detected description: >- A detection result that describes a match against licensed code or other protected source material. ungrounded_material: $ref: >- #/components/schemas/AzureContentFilterCompletionTextSpanDetectionResult personally_identifiable_information: allOf: - $ref: >- #/components/schemas/AzureContentFilterPersonallyIdentifiableInformationResult description: >- A detection result that describes matches against Personal Identifiable Information with configurable subcategories. description: >- A content filter result for a single response item produced by a generative AI system. AzureChatCompletionResponseMessage: type: object required: - content - refusal - role properties: content: type: string nullable: true description: The contents of the message. refusal: type: string nullable: true description: The refusal message generated by the model. tool_calls: $ref: '#/components/schemas/ChatCompletionMessageToolCallsItem' readOnly: true annotations: type: array items: type: object properties: type: type: string enum: - url_citation description: The type of the URL citation. Always `url_citation`. url_citation: type: object properties: end_index: type: integer format: int32 description: >- The index of the last character of the URL citation in the message. start_index: type: integer format: int32 description: >- The index of the first character of the URL citation in the message. url: type: string format: uri description: The URL of the web resource. title: type: string description: The title of the web resource. required: - end_index - start_index - url - title description: A URL citation when using web search. required: - type - url_citation description: |- Annotations for the message, when applicable, as when using the [web search tool](/docs/guides/tools-web-search?api-mode=chat). role: type: string enum: - assistant description: The role of the author of this message. function_call: type: object properties: name: type: string arguments: type: string required: - name - arguments description: >- Deprecated and replaced by `tool_calls`. The name and arguments of a function that should be called, as generated by the model. deprecated: true audio: type: object properties: id: type: string description: Unique identifier for this audio response. expires_at: type: integer format: unixtime description: >- The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations. data: type: string format: base64 description: |- Base64 encoded audio bytes generated by the model, in the format specified in the request. transcript: type: string description: Transcript of the audio generated by the model. required: - id - expires_at - data - transcript nullable: true description: >- If the audio output modality is requested, this object contains data about the audio response from the model. [Learn more](/docs/guides/audio). context: allOf: - $ref: '#/components/schemas/AzureChatMessageContext' description: >- The Azure-specific context information associated with the chat completion response message. reasoning_content: type: string description: >- An Azure-specific extension property containing generated reasoning content from supported models. description: >- The extended response model component for chat completion response messages on the Azure OpenAI service. This model adds support for chat message context, used by the On Your Data feature for intent, citations, and other information related to retrieval-augmented generation performed. AzureContentFilterResultForPrompt: type: object properties: prompt_index: type: integer format: int32 description: >- The index of the input prompt associated with the accompanying content filter result categories. content_filter_results: type: object properties: sexual: allOf: - $ref: '#/components/schemas/AzureContentFilterSeverityResult' description: >- A content filter category for language related to anatomical organs and genitals, romantic relationships, acts portrayed in erotic or affectionate terms, pregnancy, physical sexual acts, including those portrayed as an assault or a forced sexual violent act against one's will, prostitution, pornography, and abuse. hate: allOf: - $ref: '#/components/schemas/AzureContentFilterSeverityResult' description: >- A content filter category that can refer to any content that attacks or uses pejorative or discriminatory language with reference to a person or identity group based on certain differentiating attributes of these groups including but not limited to race, ethnicity, nationality, gender identity and expression, sexual orientation, religion, immigration status, ability status, personal appearance, and body size. violence: allOf: - $ref: '#/components/schemas/AzureContentFilterSeverityResult' description: >- A content filter category for language related to physical actions intended to hurt, injure, damage, or kill someone or something; describes weapons, guns and related entities, such as manufactures, associations, legislation, and so on. self_harm: allOf: - $ref: '#/components/schemas/AzureContentFilterSeverityResult' description: >- A content filter category that describes language related to physical actions intended to purposely hurt, injure, damage one's body or kill oneself. profanity: allOf: - $ref: '#/components/schemas/AzureContentFilterDetectionResult' description: >- A detection result that identifies whether crude, vulgar, or otherwise objection language is present in the content. custom_blocklists: allOf: - $ref: '#/components/schemas/AzureContentFilterBlocklistResult' description: >- A collection of binary filtering outcomes for configured custom blocklists. custom_topics: allOf: - $ref: '#/components/schemas/AzureContentFilterCustomTopicResult' description: >- A collection of binary filtering outcomes for configured custom topics. error: type: object properties: code: type: integer format: int32 description: A distinct, machine-readable code associated with the error. message: type: string description: A human-readable message associated with the error. required: - code - message description: >- If present, details about an error that prevented content filtering from completing its evaluation. jailbreak: allOf: - $ref: '#/components/schemas/AzureContentFilterDetectionResult' description: >- A detection result that describes user prompt injection attacks, where malicious users deliberately exploit system vulnerabilities to elicit unauthorized behavior from the LLM. This could lead to inappropriate content generation or violations of system-imposed restrictions. indirect_attack: allOf: - $ref: '#/components/schemas/AzureContentFilterDetectionResult' description: >- A detection result that describes attacks on systems powered by Generative AI models that can happen every time an application processes information that wasn’t directly authored by either the developer of the application or the user. required: - jailbreak - indirect_attack description: The content filter category details for the result. description: >- A content filter result associated with a single input prompt item into a generative AI system. OpenAI.ChatCompletionStreamResponseDelta: type: object properties: audio: allOf: - $ref: '#/components/schemas/OpenAI.ChatCompletionMessageAudioChunk' description: Response audio associated with the streaming chat delta payload. content: type: string nullable: true description: The contents of the chunk message. function_call: type: object properties: name: type: string arguments: type: string description: >- Deprecated and replaced by `tool_calls`. The name and arguments of a function that should be called, as generated by the model. deprecated: true tool_calls: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionMessageToolCallChunk' readOnly: true role: allOf: - $ref: '#/components/schemas/OpenAI.ChatCompletionRole' description: The role of the author of this message. refusal: type: string nullable: true description: The refusal message generated by the model. description: A chat completion delta generated by streamed model responses. AzureChatCompletionStreamResponseDelta: type: object properties: audio: allOf: - $ref: '#/components/schemas/OpenAI.ChatCompletionMessageAudioChunk' description: Response audio associated with the streaming chat delta payload. content: type: string nullable: true description: The contents of the chunk message. function_call: type: object properties: name: type: string arguments: type: string description: >- Deprecated and replaced by `tool_calls`. The name and arguments of a function that should be called, as generated by the model. deprecated: true tool_calls: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionMessageToolCallChunk' readOnly: true role: allOf: - $ref: '#/components/schemas/OpenAI.ChatCompletionRole' description: The role of the author of this message. refusal: type: string nullable: true description: The refusal message generated by the model. context: allOf: - $ref: '#/components/schemas/AzureChatMessageContext' description: >- The Azure-specific context information associated with the chat completion response message. reasoning_content: type: string description: >- An Azure-specific extension property containing generated reasoning content from supported models. description: >- The extended response model for a streaming chat response message on the Azure OpenAI service. This model adds support for chat message context, used by the On Your Data feature for intent, citations, and other information related to retrieval-augmented generation performed. OpenAI.ChatCompletionRole: type: string enum: - system - developer - user - assistant - tool - function description: The role of the author of a message OpenAI.ChatCompletionRequestMessageContentPart: type: object required: - type properties: type: $ref: >- #/components/schemas/OpenAI.ChatCompletionRequestMessageContentPartType discriminator: propertyName: type mapping: text: >- #/components/schemas/OpenAI.ChatCompletionRequestMessageContentPartText image_url: >- #/components/schemas/OpenAI.ChatCompletionRequestMessageContentPartImage refusal: >- #/components/schemas/OpenAI.ChatCompletionRequestMessageContentPartRefusal file: >- #/components/schemas/OpenAI.ChatCompletionRequestMessageContentPartFile input_audio: >- #/components/schemas/OpenAI.ChatCompletionRequestMessageContentPartAudio OpenAI.ChatOutputPredictionType: anyOf: - type: string - type: string enum: - content OpenAI.FunctionObject: type: object required: - name properties: description: type: string description: >- A description of what the function does, used by the model to choose when and how to call the function. name: type: string description: >- The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64. parameters: description: >- The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format. Omitting `parameters` defines a function with an empty parameter list. strict: type: boolean nullable: true description: >- Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the `parameters` field. Only a subset of JSON Schema is supported when `strict` is `true`. Learn more about Structured Outputs in the [function calling guide](docs/guides/function-calling). default: false OpenAI.ChatCompletionNamedToolChoice: type: object required: - type - function properties: type: type: string enum: - function description: The type of the tool. Currently, only `function` is supported. function: type: object properties: name: type: string description: The name of the function to call. required: - name description: >- Specifies a tool the model should use. Use to force the model to call a specific function. AzureChatDataSourceType: type: string enum: - azure_search - azure_cosmos_db - elasticsearch - pinecone - mongo_db AzureContentFilterSeverityResult: type: object required: - filtered - severity properties: filtered: type: boolean description: Whether the content severity resulted in a content filtering action. severity: type: string enum: - safe - low - medium - high description: The labeled severity of the content. description: >- A labeled content filter result item that indicates whether the content was filtered and what the qualitative severity level of the content was, as evaluated against content filter configuration for the category. AzureContentFilterDetectionResult: type: object required: - filtered - detected properties: filtered: type: boolean description: >- Whether the content detection resulted in a content filtering action. detected: type: boolean description: Whether the labeled content category was detected in the content. description: >- A labeled content filter result item that indicates whether the content was detected and whether the content was filtered. AzureContentFilterBlocklistResult: type: object required: - filtered properties: filtered: type: boolean description: >- A value indicating whether any of the detailed blocklists resulted in a filtering action. details: type: array items: type: object properties: filtered: type: boolean description: >- A value indicating whether the blocklist produced a filtering action. id: type: string description: The ID of the custom blocklist evaluated. required: - filtered - id description: >- The pairs of individual blocklist IDs and whether they resulted in a filtering action. description: >- A collection of true/false filtering results for configured custom blocklists. AzureContentFilterCustomTopicResult: type: object required: - filtered properties: filtered: type: boolean description: >- A value indicating whether any of the detailed topics resulted in a filtering action. details: type: array items: type: object properties: detected: type: boolean description: A value indicating whether the topic is detected. id: type: string description: The ID of the custom topic evaluated. required: - detected - id description: The pairs of individual topic IDs and whether they are detected. description: >- A collection of true/false filtering results for configured custom topics. AzureContentFilterCompletionTextSpanDetectionResult: type: object required: - filtered - detected - details properties: filtered: type: boolean description: >- Whether the content detection resulted in a content filtering action. detected: type: boolean description: Whether the labeled content category was detected in the content. details: type: array items: $ref: '#/components/schemas/AzureContentFilterCompletionTextSpan' description: Detailed information about the detected completion text spans. AzureContentFilterPersonallyIdentifiableInformationResult: type: object properties: redacted_text: type: string description: The redacted text with PII information removed or masked. sub_categories: type: array items: $ref: '#/components/schemas/AzurePiiSubCategoryResult' description: Detailed results for individual PIIHarmSubCategory(s). allOf: - $ref: '#/components/schemas/AzureContentFilterDetectionResult' description: >- A content filter detection result for Personally Identifiable Information that includes harm extensions. ChatCompletionMessageToolCallsItem: type: array items: $ref: '#/components/schemas/OpenAI.ChatCompletionMessageToolCall' description: The tool calls generated by the model, such as function calls. AzureChatMessageContext: type: object properties: intent: type: string description: >- The detected intent from the chat history, which is used to carry conversation context between interactions citations: type: array items: type: object properties: content: type: string description: The content of the citation. title: type: string description: The title for the citation. url: type: string description: The URL of the citation. filepath: type: string description: The file path for the citation. chunk_id: type: string description: The chunk ID for the citation. rerank_score: type: number format: double description: The rerank score for the retrieval. required: - content description: The citations produced by the data retrieval. all_retrieved_documents: type: object properties: content: type: string description: The content of the citation. title: type: string description: The title for the citation. url: type: string description: The URL of the citation. filepath: type: string description: The file path for the citation. chunk_id: type: string description: The chunk ID for the citation. rerank_score: type: number format: double description: The rerank score for the retrieval. search_queries: type: array items: type: string description: The search queries executed to retrieve documents. data_source_index: type: integer format: int32 description: The index of the data source used for retrieval. original_search_score: type: number format: double description: The original search score for the retrieval. filter_reason: type: string enum: - score - rerank description: If applicable, an indication of why the document was filtered. required: - content - search_queries - data_source_index description: >- Summary information about documents retrieved by the data retrieval operation. description: >- An additional property, added to chat completion response messages, produced by the Azure OpenAI service when using extension behavior. This includes intent and citation information from the On Your Data feature. OpenAI.ChatCompletionMessageAudioChunk: type: object properties: id: type: string transcript: type: string data: type: string format: base64 expires_at: type: integer format: unixtime OpenAI.ChatCompletionMessageToolCallChunk: type: object required: - index properties: index: type: integer format: int32 id: type: string description: The ID of the tool call. type: type: string enum: - function description: The type of the tool. Currently, only `function` is supported. function: type: object properties: name: type: string description: The name of the function to call. arguments: type: string description: >- The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function. OpenAI.ChatCompletionRequestMessageContentPartType: type: string enum: - text - file - input_audio - image_url - refusal AzureContentFilterCompletionTextSpan: type: object required: - completion_start_offset - completion_end_offset properties: completion_start_offset: type: integer format: int32 description: Offset of the UTF32 code point which begins the span. completion_end_offset: type: integer format: int32 description: >- Offset of the first UTF32 code point which is excluded from the span. This field is always equal to completion_start_offset for empty spans. This field is always larger than completion_start_offset for non-empty spans. description: >- A representation of a span of completion text as used by Azure OpenAI content filter results. AzurePiiSubCategoryResult: type: object required: - sub_category - filtered - detected - redacted properties: sub_category: type: string description: The PIIHarmSubCategory that was evaluated. filtered: type: boolean description: >- Whether the content detection resulted in a content filtering action for this subcategory. detected: type: boolean description: Whether the labeled content subcategory was detected in the content. redacted: type: boolean description: Whether the content was redacted for this subcategory. description: Result details for individual PIIHarmSubCategory(s). OpenAI.ChatCompletionMessageToolCall: type: object required: - id - type - function properties: id: type: string description: The ID of the tool call. type: type: string enum: - function description: The type of the tool. Currently, only `function` is supported. function: type: object properties: name: type: string description: The name of the function to call. arguments: type: string description: >- The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function. required: - name - arguments description: The function that the model called. securitySchemes: ApiKeyAuth: type: apiKey in: header name: api-key ApiKeyAuth_: type: apiKey in: header name: authorization OAuth2Auth: type: oauth2 flows: implicit: authorizationUrl: https://login.microsoftonline.com/common/oauth2/v2.0/authorize scopes: https://cognitiveservices.azure.com/.default: '' ````