[Chat Completions] Request payload definition #21

[jihyunmoon16]

CreateChatCompletionRequest

참고 링크: https://github.com/Azure/azure-rest-api-specs/blob/main/specification/cognitiveservices/data-plane/AzureOpenAI/inference/stable/2024-06-01/inference.json

---

CreateChatCompletionRequest

"createChatCompletionRequest": {
        "type": "object",
        "allOf": [
          {
            "$ref": "#/components/schemas/chatCompletionsRequestCommon"
          },
          {
            "properties": {
              "messages": {
                "description": "A list of messages comprising the conversation so far. [Example Python code](https://github.com/openai/openai-cookbook/blob/main/examples/How_to_format_inputs_to_ChatGPT_models.ipynb).",
                "type": "array",
                "minItems": 1,
                "items": {
                  "$ref": "#/components/schemas/chatCompletionRequestMessage"
                }
              },
              "data_sources": {
                "type": "array",
                "description": "  The configuration entries for Azure OpenAI chat extensions that use them.\n  This additional specification is only compatible with Azure OpenAI.",
                "items": {
                  "$ref": "#/components/schemas/azureChatExtensionConfiguration"
                }
              },
              "n": {
                "type": "integer",
                "minimum": 1,
                "maximum": 128,
                "default": 1,
                "example": 1,
                "nullable": true,
                "description": "How many chat completion choices to generate for each input message."
              },
              "seed": {
                "type": "integer",
                "minimum": -9223372036854775808,
                "maximum": 9223372036854775807,
                "default": 0,
                "example": 1,
                "nullable": true,
                "description": "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": {
                "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`. This option is currently not available on the `gpt-4-vision-preview` model.",
                "type": "boolean",
                "default": false,
                "nullable": true
              },
              "top_logprobs": {
                "description": "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": "integer",
                "minimum": 0,
                "maximum": 5,
                "nullable": true
              },
              "response_format": {
                "type": "object",
                "description": "An object specifying the format that the model must output. Used to enable JSON mode.",
                "properties": {
                  "type": {
                    "$ref": "#/components/schemas/chatCompletionResponseFormat"
                  }
                }
              },
              "tools": {
                "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.",
                "type": "array",
                "minItems": 1,
                "items": {
                  "$ref": "#/components/schemas/chatCompletionTool"
                }
              },
              "tool_choice": {
                "$ref": "#/components/schemas/chatCompletionToolChoiceOption"
              },
              "functions": {
                "description": "Deprecated in favor of `tools`. A list of functions the model may generate JSON inputs for.",
                "type": "array",
                "minItems": 1,
                "maxItems": 128,
                "items": {
                  "$ref": "#/components/schemas/chatCompletionFunction"
                }
              },
              "function_call": {
                "description": "Deprecated in favor of `tool_choice`. Controls how the model responds to function calls. \"none\" means the model does not call a function, and responds to the end-user. \"auto\" means the model can pick between an end-user 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.",
                "oneOf": [
                  {
                    "type": "string",
                    "enum": [
                      "none",
                      "auto"
                    ],
                    "description": "`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."
                  },
                  {
                    "type": "object",
                    "description": "Specifying a particular function via `{\"name\": \"my_function\"}` forces the model to call that function.",
                    "properties": {
                      "name": {
                        "type": "string",
                        "description": "The name of the function to call."
                      }
                    },
                    "required": [
                      "name"
                    ]
                  }
                ]
              }
            }
          }
        ],
        "required": [
          "messages"
        ]
      },

ChatCompletionsRequestCommon

      "chatCompletionsRequestCommon": {
        "type": "object",
        "properties": {
          "temperature": {
            "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.\nWe generally recommend altering this or `top_p` but not both.",
            "type": "number",
            "minimum": 0,
            "maximum": 2,
            "default": 1,
            "example": 1,
            "nullable": true
          },
          "top_p": {
            "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.\nWe generally recommend altering this or `temperature` but not both.",
            "type": "number",
            "minimum": 0,
            "maximum": 1,
            "default": 1,
            "example": 1,
            "nullable": true
          },
          "stream": {
            "description": "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.",
            "type": "boolean",
            "nullable": true,
            "default": false
          },
          "stop": {
            "description": "Up to 4 sequences where the API will stop generating further tokens.",
            "oneOf": [
              {
                "type": "string",
                "nullable": true
              },
              {
                "type": "array",
                "items": {
                  "type": "string",
                  "nullable": false
                },
                "minItems": 1,
                "maxItems": 4,
                "description": "Array minimum size of 1 and maximum of 4"
              }
            ],
            "default": null
          },
          "max_tokens": {
            "description": "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).",
            "type": "integer",
            "default": 4096
          },
          "presence_penalty": {
            "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.",
            "type": "number",
            "default": 0,
            "minimum": -2,
            "maximum": 2
          },
          "frequency_penalty": {
            "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.",
            "type": "number",
            "default": 0,
            "minimum": -2,
            "maximum": 2
          },
          "logit_bias": {
            "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.",
            "type": "object",
            "nullable": true
          },
          "user": {
            "description": "A unique identifier representing your end-user, which can help Azure OpenAI to monitor and detect abuse.",
            "type": "string",
            "example": "user-1234",
            "nullable": false
          }
        }
      },

ChatCompletionRequestMessage

      "chatCompletionRequestMessage": {
        "type": "object",
        "properties": {
          "role": {
            "$ref": "#/components/schemas/chatCompletionRequestMessageRole"
          }
        },
        "discriminator": {
          "propertyName": "role",
          "mapping": {
            "system": "#/components/schemas/chatCompletionRequestMessageSystem",
            "user": "#/components/schemas/chatCompletionRequestMessageUser",
            "assistant": "#/components/schemas/chatCompletionRequestMessageAssistant",
            "tool": "#/components/schemas/chatCompletionRequestMessageTool",
            "function": "#/components/schemas/chatCompletionRequestMessageFunction"
          }
        },
        "required": [
          "role"
        ]
      },

ChatCompletionRequestMessageRole

      "chatCompletionRequestMessageRole": {
        "type": "string",
        "enum": [
          "system",
          "user",
          "assistant",
          "tool",
          "function"
        ],
        "description": "The role of the messages author.",
        "x-ms-enum": {
          "name": "ChatCompletionRequestMessageRole",
          "modelAsString": true,
          "values": [
            {
              "value": "system",
              "description": "The message author role is system."
            },
            {
              "value": "user",
              "description": "The message author role is user."
            },
            {
              "value": "assistant",
              "description": "The message author role is assistant."
            },
            {
              "value": "tool",
              "description": "The message author role is tool."
            },
            {
              "value": "function",
              "description": "Deprecated. The message author role is function."
            }
          ]
        }
      },

ChatCompletionRequestMessageSystem

      "chatCompletionRequestMessageSystem": {
        "allOf": [
          {
            "$ref": "#/components/schemas/chatCompletionRequestMessage"
          },
          {
            "type": "object",
            "properties": {
              "content": {
                "type": "string",
                "description": "The contents of the message.",
                "nullable": true
              }
            }
          }
        ],
        "required": [
          "content"
        ]
      },

ChatCompletionRequestMessageUser

   "chatCompletionRequestMessageUser": {
        "allOf": [
          {
            "$ref": "#/components/schemas/chatCompletionRequestMessage"
          },
          {
            "type": "object",
            "properties": {
              "content": {
                "oneOf": [
                  {
                    "type": "string",
                    "description": "The contents of the message."
                  },
                  {
                    "type": "array",
                    "description": "An array of content parts with a defined type, each can be of type `text` or `image_url` when passing in images. You can pass multiple images by adding multiple `image_url` content parts. Image input is only supported when using the `gpt-4-visual-preview` model.",
                    "minimum": 1,
                    "items": {
                      "$ref": "#/components/schemas/chatCompletionRequestMessageContentPart"
                    }
                  }
                ],
                "nullable": true
              }
            }
          }
        ],
        "required": [
          "content"
        ]
      },

ChatCompletionRequestMessageContentPart

      "chatCompletionRequestMessageContentPart": {
        "type": "object",
        "properties": {
          "type": {
            "$ref": "#/components/schemas/chatCompletionRequestMessageContentPartType"
          }
        },
        "discriminator": {
          "propertyName": "type",
          "mapping": {
            "text": "#/components/schemas/chatCompletionRequestMessageContentPartText",
            "image_url": "#/components/schemas/chatCompletionRequestMessageContentPartImage"
          }
        },
        "required": [
          "type"
        ]
      },

ChatCompletionRequestMessageAssistant

 "chatCompletionRequestMessageAssistant": {
        "allOf": [
          {
            "$ref": "#/components/schemas/chatCompletionRequestMessage"
          },
          {
            "type": "object",
            "properties": {
              "content": {
                "type": "string",
                "description": "The contents of the message.",
                "nullable": true
              },
              "tool_calls": {
                "type": "array",
                "description": "The tool calls generated by the model, such as function calls.",
                "items": {
                  "$ref": "#/components/schemas/chatCompletionMessageToolCall"
                }
              },
              "context": {
                "$ref": "#/components/schemas/azureChatExtensionsMessageContext"
              }
            }
          }
        ],
        "required": [
          "content"
        ]
      },

ChatCompletionRequestMessageTool

 "chatCompletionRequestMessageTool": {
        "allOf": [
          {
            "$ref": "#/components/schemas/chatCompletionRequestMessage"
          },
          {
            "type": "object",
            "nullable": true,
            "properties": {
              "tool_call_id": {
                "type": "string",
                "description": "Tool call that this message is responding to."
              },
              "content": {
                "type": "string",
                "description": "The contents of the message.",
                "nullable": true
              }
            }
          }
        ],
        "required": [
          "tool_call_id",
          "content"
        ]
      },

AzureChatExtensionConfiguration

      "azureChatExtensionConfiguration": {
        "required": [
          "type"
        ],
        "type": "object",
        "properties": {
          "type": {
            "$ref": "#/components/schemas/azureChatExtensionType"
          }
        },
        "description": "  A representation of configuration data for a single Azure OpenAI chat extension. This will be used by a chat\n  completions request that should use Azure OpenAI chat extensions to augment the response behavior.\n  The use of this configuration is compatible only with Azure OpenAI.",
        "discriminator": {
          "propertyName": "type",
          "mapping": {
            "azure_search": "#/components/schemas/azureSearchChatExtensionConfiguration",
            "azure_cosmos_db": "#/components/schemas/azureCosmosDBChatExtensionConfiguration"
          }
        }
      },

AzureChatExtensionType

      "azureChatExtensionType": {
        "type": "string",
        "description": "  A representation of configuration data for a single Azure OpenAI chat extension. This will be used by a chat\n  completions request that should use Azure OpenAI chat extensions to augment the response behavior.\n  The use of this configuration is compatible only with Azure OpenAI.",
        "enum": [
          "azure_search",
          "azure_cosmos_db"
        ],
        "x-ms-enum": {
          "name": "AzureChatExtensionType",
          "modelAsString": true,
          "values": [
            {
              "name": "azureSearch",
              "value": "azure_search",
              "description": "Represents the use of Azure Search as an Azure OpenAI chat extension."
            },
            {
              "name": "azureCosmosDB",
              "value": "azure_cosmos_db",
              "description": "Represents the use of Azure Cosmos DB as an Azure OpenAI chat extension."
            }
          ]
        }
      },

AzureSearchChatExtensionConfiguration

      "azureSearchChatExtensionConfiguration": {
        "required": [
          "parameters"
        ],
        "description": "A specific representation of configurable options for Azure Search when using it as an Azure OpenAI chat\nextension.",
        "allOf": [
          {
            "$ref": "#/components/schemas/azureChatExtensionConfiguration"
          },
          {
            "properties": {
              "parameters": {
                "$ref": "#/components/schemas/azureSearchChatExtensionParameters"
              }
            }
          }
        ],
        "x-ms-discriminator-value": "azure_search"
      },

AzureCosmosDBChatExtensionConfiguration

      "azureCosmosDBChatExtensionConfiguration": {
        "required": [
          "parameters"
        ],
        "description": "A specific representation of configurable options for Azure Cosmos DB when using it as an Azure OpenAI chat\nextension.",
        "allOf": [
          {
            "$ref": "#/components/schemas/azureChatExtensionConfiguration"
          },
          {
            "properties": {
              "parameters": {
                "$ref": "#/components/schemas/azureCosmosDBChatExtensionParameters"
              }
            }
          }
        ],
        "x-ms-discriminator-value": "azure_cosmos_db"
      },

AzureCosmosDBChatExtensionParameters

  "azureCosmosDBChatExtensionParameters": {
        "required": [
          "authentication",
          "container_name",
          "database_name",
          "embedding_dependency",
          "fields_mapping",
          "index_name"
        ],
        "type": "object",
        "properties": {
          "authentication": {
            "$ref": "#/components/schemas/onYourDataConnectionStringAuthenticationOptions"
          },
          "top_n_documents": {
            "type": "integer",
            "description": "The configured top number of documents to feature for the configured query.",
            "format": "int32"
          },
          "in_scope": {
            "type": "boolean",
            "description": "Whether queries should be restricted to use of indexed data."
          },
          "strictness": {
            "maximum": 5,
            "minimum": 1,
            "type": "integer",
            "description": "The configured strictness of the search relevance filtering. The higher of strictness, the higher of the precision but lower recall of the answer.",
            "format": "int32"
          },
          "role_information": {
            "type": "string",
            "description": "Give the model instructions about how it should behave and any context it should reference when generating a response. You can describe the assistant's personality and tell it how to format responses. There's a 100 token limit for it, and it counts against the overall token limit."
          },
          "database_name": {
            "type": "string",
            "description": "The MongoDB vCore database name to use with Azure Cosmos DB."
          },
          "container_name": {
            "type": "string",
            "description": "The name of the Azure Cosmos DB resource container."
          },
          "index_name": {
            "type": "string",
            "description": "The MongoDB vCore index name to use with Azure Cosmos DB."
          },
          "fields_mapping": {
            "$ref": "#/components/schemas/azureCosmosDBFieldMappingOptions"
          },
          "embedding_dependency": {
            "oneOf": [
              {
                "$ref": "#/components/schemas/onYourDataEndpointVectorizationSource"
              },
              {
                "$ref": "#/components/schemas/onYourDataDeploymentNameVectorizationSource"
              }
            ]
          }
        },
        "description": "Parameters to use when configuring Azure OpenAI On Your Data chat extensions when using Azure Cosmos DB for\nMongoDB vCore."
      },

OnYourDataConnectionStringAuthenticationOptions

      "onYourDataConnectionStringAuthenticationOptions": {
        "required": [
          "connection_string"
        ],
        "description": "The authentication options for Azure OpenAI On Your Data when using a connection string.",
        "allOf": [
          {
            "$ref": "#/components/schemas/onYourDataAuthenticationOptions"
          },
          {
            "properties": {
              "connection_string": {
                "type": "string",
                "description": "The connection string to use for authentication."
              }
            }
          }
        ],
        "x-ms-discriminator-value": "connection_string"
      },

OnYourDataAuthenticationOptions

      "onYourDataAuthenticationOptions": {
        "required": [
          "type"
        ],
        "type": "object",
        "properties": {
          "type": {
            "$ref": "#/components/schemas/onYourDataAuthenticationType"
          }
        },
        "description": "The authentication options for Azure OpenAI On Your Data.",
        "discriminator": {
          "propertyName": "type",
          "mapping": {
            "api_key": "#/components/schemas/onYourDataApiKeyAuthenticationOptions",
            "connection_string": "#/components/schemas/onYourDataConnectionStringAuthenticationOptions",
            "system_assigned_managed_identity": "#/components/schemas/onYourDataSystemAssignedManagedIdentityAuthenticationOptions",
            "user_assigned_managed_identity": "#/components/schemas/onYourDataUserAssignedManagedIdentityAuthenticationOptions"
          }
        }
      },

ChatCompletionResponseFormat

 "chatCompletionResponseFormat": {
        "type": "string",
        "enum": [
          "text",
          "json_object"
        ],
        "default": "text",
        "example": "json_object",
        "nullable": true,
        "description": "Setting to `json_object` enables JSON mode. This guarantees that the message the model generates is valid JSON.",
        "x-ms-enum": {
          "name": "ChatCompletionResponseFormat",
          "modelAsString": true,
          "values": [
            {
              "value": "text",
              "description": "Response format is a plain text string."
            },
            {
              "value": "json_object",
              "description": "Response format is a JSON object."
            }
          ]
        }
      },

ChatCompletionTool

   "chatCompletionTool": {
        "type": "object",
        "properties": {
          "type": {
            "$ref": "#/components/schemas/chatCompletionToolType"
          },
          "function": {
            "type": "object",
            "properties": {
              "description": {
                "type": "string",
                "description": "A description of what the function does, used by the model to choose when and how to call the function."
              },
              "name": {
                "type": "string",
                "description": "The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."
              },
              "parameters": {
                "$ref": "#/components/schemas/chatCompletionFunctionParameters"
              }
            },
            "required": [
              "name",
              "parameters"
            ]
          }
        },
        "required": [
          "type",
          "function"
        ]
      },

ChatCompletionToolType

     "chatCompletionToolType": {
        "type": "string",
        "enum": [
          "function"
        ],
        "description": "The type of the tool. Currently, only `function` is supported.",
        "x-ms-enum": {
          "name": "ChatCompletionToolType",
          "modelAsString": true,
          "values": [
            {
              "value": "function",
              "description": "The tool type is function."
            }
          ]
        }
      },

ChatCompletionFunctionParameters

  "chatCompletionFunctionParameters": {
        "type": "object",
        "description": "The parameters the functions accepts, described as a JSON Schema object. See the [guide](/docs/guides/gpt/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.",
        "additionalProperties": true
      },

ChatCompletionToolChoiceOption

      "chatCompletionToolChoiceOption": {
        "description": "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 `{\"type\": \"function\", \"function\": {\"name\": \"my_function\"}}` forces the model to call that function.",
        "oneOf": [
          {
            "type": "string",
            "description": "`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.",
            "enum": [
              "none",
              "auto",
              "required"
            ]
          },
          {
            "$ref": "#/components/schemas/chatCompletionNamedToolChoice"
          }
        ]
      },

ChatCompletionNamedToolChoice

      "chatCompletionNamedToolChoice": {
        "type": "object",
        "description": "Specifies a tool the model should use. Use to force the model to call a specific 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"
            ]
          }
        }
      },

[justinyoo]

@jihyunmoon16 아까 낮에 논의한 것 처럼

• ChatCompletionTool 개체의 function 어트리뷰트는 별도의 클래스로 만들어 보세요.
• 클래스 👉 JSON 직렬화 테스트 검토해 주세요.
• JSON 👉 클래스 비직렬화 테스트 검토해 주세요.

[jihyunmoon16]

테스트 코드 작성했는데 data_sources 필드는 다형성? 부분도 고려해야 할 것 같고 어떤식으로 클래스를 만들어야 할 지 고민입니다. 테스트도 계속 실패해요. 웹사이트의 examples json 파일을 참고해서 제일 기본인 required 필드만 적용해봤습니다.

코드 구현이 힘든 부분이 이 data sources 파라미터예요.

  "body": {
      "messages": [
        {
          "role": "user",
          "content": "can you tell me how to care for a cat?"
        },
        {
          "role": "assistant",
          "content": "Content of the completion [doc1].",
          "context": {
            "intent": "cat care"
          }
        },
        {
          "role": "user",
          "content": "how about dog?"
        }
      ],
      "data_sources": [
        {
          "type": "azure_search",
          "parameters": {
            "endpoint": "https://your-search-endpoint.search.windows.net/",
            "authentication": {
              "type": "user_assigned_managed_identity",
              "managed_identity_resource_id": "/subscriptions/{subscription-id}/resourceGroups/{resource-group}/providers/Microsoft.ManagedIdentity/userAssignedIdentities/{resource-name}"
            },
            "index_name": "{index name}",
            "query_type": "vector",
            "embedding_dependency": {
              "type": "deployment_name",
              "deployment_name": "{embedding deployment name}"
            },
            "in_scope": true,
            "top_n_documents": 5,
            "strictness": 3,
            "role_information": "You are an AI assistant that helps people find information.",
            "fields_mapping": {
              "content_fields_separator": "\\n",
              "content_fields": [
                "content"
              ],
              "filepath_field": "filepath",
              "title_field": "title",
              "url_field": "url",
              "vector_fields": [
                "contentvector"
              ]
            }
          }
        }
      ]
    }

테스트 코드를 작성하다가 궁금한 부분이 있는데 createChatCompletionRequest의 경우 request payload 여서 비직렬화만 할 것 같은데 이때도 직렬화 테스트가 필요할까요?