generated.json

{
  "swagger": "2.0",
  "info": {
    "title": "Azure OpenAI API",
    "version": "2022-12-01",
    "description": "Azure OpenAI APIs for completions and search",
    "x-typespec-generated": [
      {
        "emitter": "@azure-tools/typespec-autorest"
      }
    ]
  },
  "schemes": [
    "https"
  ],
  "x-ms-parameterized-host": {
    "hostTemplate": "{endpoint}/openai",
    "useSchemePrefix": false,
    "parameters": [
      {
        "name": "endpoint",
        "in": "path",
        "description": "Supported Cognitive Services endpoints (protocol and hostname, for example:\nhttps://westus.api.cognitive.microsoft.com).",
        "required": true,
        "type": "string"
      }
    ]
  },
  "produces": [
    "application/json"
  ],
  "consumes": [
    "application/json"
  ],
  "security": [
    {
      "ApiKeyAuth": []
    },
    {
      "OAuth2Auth": [
        "https://cognitiveservices.azure.com/.default"
      ]
    }
  ],
  "securityDefinitions": {
    "ApiKeyAuth": {
      "type": "apiKey",
      "name": "api-key",
      "in": "header"
    },
    "OAuth2Auth": {
      "type": "oauth2",
      "flow": "implicit",
      "authorizationUrl": "https://login.microsoftonline.com/common/oauth2/v2.0/authorize",
      "scopes": {
        "https://cognitiveservices.azure.com/.default": ""
      }
    }
  },
  "tags": [],
  "paths": {
    "/deployments/{deploymentId}/completions": {
      "post": {
        "operationId": "GetCompletions",
        "description": "Gets completions for the provided input prompts.\nCompletions support a wide variety of tasks and generate text that continues from or \"completes\"\nprovided prompt data.",
        "parameters": [
          {
            "$ref": "#/parameters/Azure.Core.Foundations.ApiVersionParameter"
          },
          {
            "name": "deploymentId",
            "in": "path",
            "description": "Specifies either the model deployment name (when using Azure OpenAI) or model name (when using non-Azure OpenAI) to use for this request.",
            "required": true,
            "type": "string"
          },
          {
            "name": "body",
            "in": "body",
            "description": "Body parameter.",
            "required": true,
            "schema": {
              "$ref": "#/definitions/CompletionsOptions"
            }
          }
        ],
        "responses": {
          "200": {
            "description": "The request has succeeded.",
            "schema": {
              "$ref": "#/definitions/Completions"
            }
          },
          "default": {
            "description": "An unexpected error response.",
            "schema": {
              "$ref": "#/definitions/Azure.Core.Foundations.ErrorResponse"
            },
            "headers": {
              "x-ms-error-code": {
                "type": "string",
                "description": "String error code indicating what went wrong."
              }
            }
          }
        },
        "x-ms-examples": {
          "Creates a completion for the provided prompt, parameters and chosen model.": {
            "$ref": "./examples/completions.json"
          }
        }
      }
    },
    "/deployments/{deploymentId}/embeddings": {
      "post": {
        "operationId": "GetEmbeddings",
        "description": "Return the embeddings for a given prompt.",
        "parameters": [
          {
            "$ref": "#/parameters/Azure.Core.Foundations.ApiVersionParameter"
          },
          {
            "name": "deploymentId",
            "in": "path",
            "description": "Specifies either the model deployment name (when using Azure OpenAI) or model name (when using non-Azure OpenAI) to use for this request.",
            "required": true,
            "type": "string"
          },
          {
            "name": "body",
            "in": "body",
            "description": "Body parameter.",
            "required": true,
            "schema": {
              "$ref": "#/definitions/EmbeddingsOptions"
            }
          }
        ],
        "responses": {
          "200": {
            "description": "The request has succeeded.",
            "schema": {
              "$ref": "#/definitions/Embeddings"
            }
          },
          "default": {
            "description": "An unexpected error response.",
            "schema": {
              "$ref": "#/definitions/Azure.Core.Foundations.ErrorResponse"
            },
            "headers": {
              "x-ms-error-code": {
                "type": "string",
                "description": "String error code indicating what went wrong."
              }
            }
          }
        },
        "x-ms-examples": {
          "Return the embeddings for a given prompt.": {
            "$ref": "./examples/embeddings.json"
          }
        }
      }
    }
  },
  "definitions": {
    "Azure.Core.Foundations.Error": {
      "type": "object",
      "description": "The error object.",
      "properties": {
        "code": {
          "type": "string",
          "description": "One of a server-defined set of error codes."
        },
        "message": {
          "type": "string",
          "description": "A human-readable representation of the error."
        },
        "target": {
          "type": "string",
          "description": "The target of the error."
        },
        "details": {
          "type": "array",
          "description": "An array of details about specific errors that led to this reported error.",
          "items": {
            "$ref": "#/definitions/Azure.Core.Foundations.Error"
          },
          "x-ms-identifiers": []
        },
        "innererror": {
          "$ref": "#/definitions/Azure.Core.Foundations.InnerError",
          "description": "An object containing more specific information than the current object about the error."
        }
      },
      "required": [
        "code",
        "message"
      ]
    },
    "Azure.Core.Foundations.ErrorResponse": {
      "type": "object",
      "description": "A response containing error details.",
      "properties": {
        "error": {
          "$ref": "#/definitions/Azure.Core.Foundations.Error",
          "description": "The error object."
        }
      },
      "required": [
        "error"
      ]
    },
    "Azure.Core.Foundations.InnerError": {
      "type": "object",
      "description": "An object containing more specific information about the error. As per Microsoft One API guidelines - https://github.com/Microsoft/api-guidelines/blob/vNext/Guidelines.md#7102-error-condition-responses.",
      "properties": {
        "code": {
          "type": "string",
          "description": "One of a server-defined set of error codes."
        },
        "innererror": {
          "$ref": "#/definitions/Azure.Core.Foundations.InnerError",
          "description": "Inner error."
        }
      }
    },
    "Choice": {
      "type": "object",
      "description": "The representation of a single prompt completion as part of an overall completions request.\nGenerally, `n` choices are generated per provided prompt with a default value of 1.\nToken limits and other settings may limit the number of choices generated.",
      "properties": {
        "text": {
          "type": "string",
          "description": "The generated text for a given completions prompt."
        },
        "index": {
          "type": "integer",
          "format": "int32",
          "description": "The ordered index associated with this completions choice."
        },
        "logprobs": {
          "type": "object",
          "description": "The log probabilities model for tokens associated with this completions choice.",
          "x-nullable": true,
          "allOf": [
            {
              "$ref": "#/definitions/CompletionsLogProbabilityModel"
            }
          ]
        },
        "finish_reason": {
          "$ref": "#/definitions/CompletionsFinishReason",
          "description": "Reason for finishing",
          "x-nullable": true,
          "x-ms-client-name": "finishReason"
        }
      },
      "required": [
        "text",
        "index",
        "logprobs",
        "finish_reason"
      ]
    },
    "Completions": {
      "type": "object",
      "description": "Representation of the response data from a completions request.\nCompletions support a wide variety of tasks and generate text that continues from or \"completes\"\nprovided prompt data.",
      "properties": {
        "id": {
          "type": "string",
          "description": "A unique identifier associated with this completions response."
        },
        "created": {
          "type": "integer",
          "format": "unixtime",
          "description": "The first timestamp associated with generation activity for this completions response,\nrepresented as seconds since the beginning of the Unix epoch of 00:00 on 1 Jan 1970."
        },
        "choices": {
          "type": "array",
          "description": "The collection of completions choices associated with this completions response.\nGenerally, `n` choices are generated per provided prompt with a default value of 1.\nToken limits and other settings may limit the number of choices generated.",
          "items": {
            "$ref": "#/definitions/Choice"
          },
          "x-ms-identifiers": []
        },
        "usage": {
          "$ref": "#/definitions/CompletionsUsage",
          "description": "Usage information for tokens processed and generated as part of this completions operation."
        }
      },
      "required": [
        "id",
        "created",
        "choices",
        "usage"
      ]
    },
    "CompletionsFinishReason": {
      "type": "string",
      "description": "Representation of the manner in which a completions response concluded.",
      "enum": [
        "stop",
        "length",
        "content_filter"
      ],
      "x-ms-enum": {
        "name": "CompletionsFinishReason",
        "modelAsString": true,
        "values": [
          {
            "name": "stopped",
            "value": "stop",
            "description": "Completions ended normally and reached its end of token generation."
          },
          {
            "name": "tokenLimitReached",
            "value": "length",
            "description": "Completions exhausted available token limits before generation could complete."
          },
          {
            "name": "contentFiltered",
            "value": "content_filter",
            "description": "Completions generated a response that was identified as potentially sensitive per content\nmoderation policies."
          }
        ]
      }
    },
    "CompletionsLogProbabilityModel": {
      "type": "object",
      "description": "Representation of a log probabilities model for a completions generation.",
      "properties": {
        "tokens": {
          "type": "array",
          "description": "The textual forms of tokens evaluated in this probability model.",
          "items": {
            "type": "string"
          }
        },
        "token_logprobs": {
          "type": "array",
          "description": "A collection of log probability values for the tokens in this completions data.",
          "items": {
            "type": "number",
            "format": "float",
            "x-nullable": true
          },
          "x-ms-client-name": "tokenLogprobs"
        },
        "top_logprobs": {
          "type": "array",
          "description": "A mapping of tokens to maximum log probability values in this completions data.",
          "items": {
            "type": "object",
            "additionalProperties": {
              "format": "float",
              "type": "number",
              "x-nullable": true
            }
          },
          "x-ms-client-name": "topLogprobs",
          "x-ms-identifiers": []
        },
        "text_offset": {
          "type": "array",
          "description": "The text offsets associated with tokens in this completions data.",
          "items": {
            "type": "integer",
            "format": "int32"
          },
          "x-ms-client-name": "textOffset"
        }
      },
      "required": [
        "tokens",
        "token_logprobs",
        "top_logprobs",
        "text_offset"
      ]
    },
    "CompletionsOptions": {
      "type": "object",
      "description": "The configuration information for a completions request.\nCompletions support a wide variety of tasks and generate text that continues from or \"completes\"\nprovided prompt data.",
      "properties": {
        "prompt": {
          "type": "array",
          "description": "The prompts to generate completions from.",
          "items": {
            "type": "string"
          }
        },
        "max_tokens": {
          "type": "integer",
          "format": "int32",
          "description": "The maximum number of tokens to generate.",
          "x-ms-client-name": "maxTokens"
        },
        "temperature": {
          "type": "number",
          "format": "float",
          "description": "The sampling temperature to use that controls the apparent creativity of generated completions.\nHigher values will make output more random while lower values will make results more focused\nand deterministic.\nIt is not recommended to modify temperature and top_p for the same completions request as the\ninteraction of these two settings is difficult to predict."
        },
        "top_p": {
          "type": "number",
          "format": "float",
          "description": "An alternative to sampling with temperature called nucleus sampling. This value causes the\nmodel to consider the results of tokens with the provided probability mass. As an example, a\nvalue of 0.15 will cause only the tokens comprising the top 15% of probability mass to be\nconsidered.\nIt is not recommended to modify temperature and top_p for the same completions request as the\ninteraction of these two settings is difficult to predict.",
          "x-ms-client-name": "topP"
        },
        "logit_bias": {
          "type": "object",
          "description": "A map between GPT token IDs and bias scores that influences the probability of specific tokens\nappearing in a completions response. Token IDs are computed via external tokenizer tools, while\nbias scores reside in the range of -100 to 100 with minimum and maximum values corresponding to\na full ban or exclusive selection of a token, respectively. The exact behavior of a given bias\nscore varies by model.",
          "additionalProperties": {
            "format": "int32",
            "type": "integer"
          },
          "x-ms-client-name": "logitBias"
        },
        "user": {
          "type": "string",
          "description": "An identifier for the caller or end user of the operation. This may be used for tracking\nor rate-limiting purposes."
        },
        "n": {
          "type": "integer",
          "format": "int32",
          "description": "The number of completions choices that should be generated per provided prompt as part of an\noverall completions response.\nBecause this setting can generate many completions, it may quickly consume your token quota.\nUse carefully and ensure reasonable settings for max_tokens and stop."
        },
        "logprobs": {
          "type": "integer",
          "format": "int32",
          "description": "A value that controls the emission of log probabilities for the provided number of most likely\ntokens within a completions response."
        },
        "suffix": {
          "type": "string",
          "description": "The suffix that comes after a completion of inserted text"
        },
        "echo": {
          "type": "boolean",
          "description": "A value specifying whether completions responses should include input prompts as prefixes to\ntheir generated output."
        },
        "stop": {
          "type": "array",
          "description": "A collection of textual sequences that will end completions generation.",
          "items": {
            "type": "string"
          }
        },
        "presence_penalty": {
          "type": "number",
          "format": "float",
          "description": "A value that influences the probability of generated tokens appearing based on their existing\npresence in generated text.\nPositive values will make tokens less likely to appear when they already exist and increase the\nmodel's likelihood to output new topics.",
          "x-ms-client-name": "presencePenalty"
        },
        "frequency_penalty": {
          "type": "number",
          "format": "float",
          "description": "A value that influences the probability of generated tokens appearing based on their cumulative\nfrequency in generated text.\nPositive values will make tokens less likely to appear as their frequency increases and\ndecrease the likelihood of the model repeating the same statements verbatim.",
          "x-ms-client-name": "frequencyPenalty"
        },
        "best_of": {
          "type": "integer",
          "format": "int32",
          "description": "A value that controls how many completions will be internally generated prior to response\nformulation.\nWhen used together with n, best_of controls the number of candidate completions and must be\ngreater than n.\nBecause this setting can generate many completions, it may quickly consume your token quota.\nUse carefully and ensure reasonable settings for max_tokens and stop.",
          "x-ms-client-name": "bestOf"
        },
        "stream": {
          "type": "boolean",
          "description": "A value indicating whether chat completions should be streamed for this request."
        },
        "model": {
          "type": "string",
          "description": "The model name to provide as part of this completions request.\nNot applicable to Azure OpenAI, where deployment information should be included in the Azure\nresource URI that's connected to."
        }
      },
      "required": [
        "prompt"
      ]
    },
    "CompletionsUsage": {
      "type": "object",
      "description": "Representation of the token counts processed for a completions request.\nCounts consider all tokens across prompts, choices, choice alternates, best_of generations, and\nother consumers.",
      "properties": {
        "completion_tokens": {
          "type": "integer",
          "format": "int32",
          "description": "The number of tokens generated across all completions emissions.",
          "x-ms-client-name": "completionTokens"
        },
        "prompt_tokens": {
          "type": "integer",
          "format": "int32",
          "description": "The number of tokens in the provided prompts for the completions request.",
          "x-ms-client-name": "promptTokens"
        },
        "total_tokens": {
          "type": "integer",
          "format": "int32",
          "description": "The total number of tokens processed for the completions request and response.",
          "x-ms-client-name": "totalTokens"
        }
      },
      "required": [
        "completion_tokens",
        "prompt_tokens",
        "total_tokens"
      ]
    },
    "EmbeddingItem": {
      "type": "object",
      "description": "Representation of a single embeddings relatedness comparison.",
      "properties": {
        "embedding": {
          "type": "array",
          "description": "List of embeddings value for the input prompt. These represent a measurement of the\nvector-based relatedness of the provided input.",
          "items": {
            "type": "number",
            "format": "float"
          }
        },
        "index": {
          "type": "integer",
          "format": "int32",
          "description": "Index of the prompt to which the EmbeddingItem corresponds."
        }
      },
      "required": [
        "embedding",
        "index"
      ]
    },
    "Embeddings": {
      "type": "object",
      "description": "Representation of the response data from an embeddings request.\nEmbeddings measure the relatedness of text strings and are commonly used for search, clustering,\nrecommendations, and other similar scenarios.",
      "properties": {
        "data": {
          "type": "array",
          "description": "Embedding values for the prompts submitted in the request.",
          "items": {
            "$ref": "#/definitions/EmbeddingItem"
          },
          "x-ms-identifiers": []
        },
        "usage": {
          "$ref": "#/definitions/EmbeddingsUsage",
          "description": "Usage counts for tokens input using the embeddings API."
        }
      },
      "required": [
        "data",
        "usage"
      ]
    },
    "EmbeddingsOptions": {
      "type": "object",
      "description": "The configuration information for an embeddings request.\nEmbeddings measure the relatedness of text strings and are commonly used for search, clustering,\nrecommendations, and other similar scenarios.",
      "properties": {
        "user": {
          "type": "string",
          "description": "An identifier for the caller or end user of the operation. This may be used for tracking\nor rate-limiting purposes."
        },
        "model": {
          "type": "string",
          "description": "The model name to provide as part of this embeddings request.\nNot applicable to Azure OpenAI, where deployment information should be included in the Azure\nresource URI that's connected to."
        },
        "input": {
          "type": "array",
          "description": "Input texts to get embeddings for, encoded as a an array of strings.\nEach input must not exceed 2048 tokens in length.\n\nUnless you are embedding code, we suggest replacing newlines (\\n) in your input with a single space,\nas we have observed inferior results when newlines are present.",
          "items": {
            "type": "string"
          }
        }
      },
      "required": [
        "input"
      ]
    },
    "EmbeddingsUsage": {
      "type": "object",
      "description": "Measurement of the amount of tokens used in this request and response.",
      "properties": {
        "prompt_tokens": {
          "type": "integer",
          "format": "int32",
          "description": "Number of tokens sent in the original request.",
          "x-ms-client-name": "promptTokens"
        },
        "total_tokens": {
          "type": "integer",
          "format": "int32",
          "description": "Total number of tokens transacted in this request/response.",
          "x-ms-client-name": "totalTokens"
        }
      },
      "required": [
        "prompt_tokens",
        "total_tokens"
      ]
    }
  },
  "parameters": {
    "Azure.Core.Foundations.ApiVersionParameter": {
      "name": "api-version",
      "in": "query",
      "description": "The API version to use for this operation.",
      "required": true,
      "type": "string",
      "minLength": 1,
      "x-ms-parameter-location": "method",
      "x-ms-client-name": "apiVersion"
    }
  }
}