Azure OpenAI in Microsoft Foundry Models authoring REST API preview reference

이 글은 Azure OpenAI용 추론 REST API 엔드포인트에 대한 자세한 내용을 제공합니다.

API specs

Azure OpenAI 모델과 자원을 관리하고 상호작용하는 것은 세 가지 주요 API 표면으로 나뉩니다:

  • Control plane
  • 데이터 평면 - 저작
  • 데이터 평면 - 추론

각 API 서피스/명세는 서로 다른 Azure OpenAI 기능 집합을 캡슐화합니다. 각 API는 고유한 미리보기와 안정적/일반 제공(GA) API 릴리스 세트를 가지고 있습니다. 현재 미리보기 릴리스는 보통 월간 일정을 따르는 경향이 있습니다.

Important

이제 새로운 미리보기 추론 API가 추가되었습니다. 자세한 내용은 API 수명주기 가이드에서 확인하세요.

API 최신 미리보기 발표 최신 GA 릴리스 Specifications Description
Control plane 2025-07-01-preview 2025-06-01 Spec files 제어 평면 API는 자원 생성, 모델 배포, 기타 상위 자원 관리 작업과 같은 작업에 사용됩니다. 제어 플레인은 또한 Azure Resource Manager, Bicep, Terraform, Azure CLI 같은 기능을 수행할 수 있는 것을 관리합니다.
Data plane v1 preview v1 Spec files 데이터 플레인 API는 추론 및 저작 작업을 제어합니다.

Authentication

Azure OpenAI는 두 가지 인증 방법을 제공합니다. API 키나 Microsoft Entra ID 중 하나를 사용할 수 있습니다.

  • API 키 인증: 이 유형의 인증에서는 모든 API 요청이 HTTP 헤더에 api-key API 키를 포함해야 합니다. 퀵스타트는 이러한 인증 방식으로 전화를 걸는 방법에 대한 안내를 제공합니다.

  • Microsoft Entra ID 인증: Microsoft Entra 토큰을 사용해 API 호출을 인증할 수 있습니다. 인증 토큰은 요청 시 헤더로 Authorization 포함됩니다. 제공된 토큰은 앞에 Bearer, 예를 Bearer YOUR_AUTH_TOKEN들어 로 로 붙어야 한다. Microsoft Entra ID로 인증하는 방법에 관한 저희 사용 가이드를 읽어보실 수 있습니다.

REST API 버전 관리

서비스 API는 쿼리 매개변수를 사용하여 api-version 버전이 지정됩니다. 모든 버전은 YYYY-MM-DD 날짜 구조를 따릅니다. For example:

POST https://YOUR_RESOURCE_NAME.openai.azure.com/openai/deployments/YOUR_DEPLOYMENT_NAME/chat/completions?api-version=2024-06-01

데이터 평면 저작

나머지 글은 Azure OpenAI 데이터 평면 추론 명세서인 2025-04-01-preview의 미리보기 릴리스를 다룹니다.

최신 GA API 릴리스에 대한 문서를 찾고 있다면, 최신 GA 데이터 플레인 추론 API를 참고하세요

배치 - 목록

GET {endpoint}/openai/batches?api-version=2025-04-01-preview

Azure OpenAI 리소스가 소유한 모든 배치 목록을 얻습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
after query No string 이전 페이지네이션 요청에서 발생한 마지막 이벤트의 식별자입니다.
limit query No integer 회수해야 할 배치의 수. 기본값은 20입니다.
$filter query No string 필터링 조건을 설명하기 위한 OData 표현식입니다.
$orderby query No string 정렬 순서를 설명하는 OData 표현식입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json BatchesList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/batches?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "has_more": false,
    "data": [
      {
        "object": "batch",
        "id": "batch_72a2792ef7d24ba7b82c7fe4a37e379f",
        "endpoint": "/chat/completions",
        "errors": null,
        "input_file_id": "file-b4fa7277f3eb4722ac18b90f46102c3f",
        "completion_window": "24h",
        "status": "completed",
        "output_file_id": "file-f2ddaf43-b48b-46dd-b264-90da10c7a85b",
        "error_file_id": "file-c3b563b0-ebc7-47da-93e3-a2fa694aef0c",
        "created_at": 1646126127,
        "in_progress_at": 1646126130,
        "expires_at": 1646126170,
        "finalizing_at": 1646126134,
        "completed_at": 1646126136,
        "failed_at": null,
        "expired_at": null,
        "cancelling_at": null,
        "cancelled_at": null,
        "request_counts": {
          "total": 500,
          "completed": 400,
          "failed": 100
        },
        "first_id": "batch_abc123",
        "last_id": "batch_abc456",
        "metadata": {
          "batch_description": "Evaluation job"
        }
      }
    ],
    "object": "list"
  }
}

배치 - 생성

POST {endpoint}/openai/batches?api-version=2025-04-01-preview

업로드된 요청 파일에서 배치를 생성하고 실행합니다.

응답에는 대기열에 걸린 작업의 세부 정보와 작업 상태가 포함됩니다.

결과 파일의 ID는 완료 후 응답에 추가됩니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: application/json

Name Type Description Required Default
completion_window string 배치가 처리되어야 하는 시간 프레임. Yes
endpoint string 배치가 사용하는 API 엔드포인트입니다. Yes
input_blob string 배치 입력으로 사용할 Azure Storage 블롭의 URL. No
input_file_id string 배치의 입력 파일의 ID입니다. No
metadata object 배치에 부착할 수 있는 키-값 쌍 집합입니다. 이는 배치에 대한 추가 정보를 구조화된 형식으로 저장하는 데 유용할 수 있습니다. No
output_expires_after FileExpiresAfter 파일의 만료일을 정의합니다. No
output_folder BatchOutputReference 출력을 저장하는 Azure Storage 폴더. No

Responses

상태 코드: 201

설명: 배치가 성공적으로 생성되었습니다.

Content-Type Type Description
application/json Batch

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/batches?api-version=2025-04-01-preview

응답: 상태 코드: 201

{
  "headers": {
    "deployment-enqueued-tokens": 1000,
    "deployment-maximum-enqueued-tokens": 5000
  },
  "body": {
    "object": "batch",
    "id": "batch_72a2792ef7d24ba7b82c7fe4a37e379f",
    "endpoint": "/chat/completions",
    "errors": null,
    "input_file_id": "file-b4fa7277f3eb4722ac18b90f46102c3f",
    "completion_window": "24h",
    "status": "validating",
    "output_file_id": null,
    "error_file_id": null,
    "created_at": 1646126127,
    "in_progress_at": null,
    "expires_at": null,
    "finalizing_at": null,
    "completed_at": null,
    "failed_at": null,
    "expired_at": null,
    "cancelling_at": null,
    "cancelled_at": null,
    "request_counts": {
      "total": 0,
      "completed": 0,
      "failed": 0
    },
    "metadata": {
      "batch_description": "Evaluation job"
    }
  }
}

배치 - 받기

GET {endpoint}/openai/batches/{batch-id}?api-version=2025-04-01-preview

주어진 배치 ID로 지정된 단일 배치의 세부 정보를 얻습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
batch-id path Yes string 배치의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json Batch

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/batches/{batch-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "batch",
    "id": "batch_72a2792ef7d24ba7b82c7fe4a37e379f",
    "endpoint": "/chat/completions",
    "errors": null,
    "input_file_id": "file-b4fa7277f3eb4722ac18b90f46102c3f",
    "completion_window": "24h",
    "status": "completed",
    "output_file_id": "file-f2ddaf43-b48b-46dd-b264-90da10c7a85b",
    "error_file_id": "file-c3b563b0-ebc7-47da-93e3-a2fa694aef0c",
    "created_at": 1646126127,
    "in_progress_at": 1646126130,
    "expires_at": 1646126170,
    "finalizing_at": 1646126134,
    "completed_at": 1646126136,
    "failed_at": null,
    "expired_at": null,
    "cancelling_at": null,
    "cancelled_at": null,
    "request_counts": {
      "total": 500,
      "completed": 400,
      "failed": 100
    },
    "metadata": {
      "batch_description": "Evaluation job"
    }
  }
}

배치 - 취소

POST {endpoint}/openai/batches/{batch-id}/cancel?api-version=2025-04-01-preview

주어진 배치 ID로 지정된 배치의 처리를 취소합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
batch-id path Yes string 배치의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 배치가 성공적으로 취소되었습니다.

Content-Type Type Description
application/json Batch

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/batches/{batch-id}/cancel?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "batch",
    "id": "batch_72a2792ef7d24ba7b82c7fe4a37e379f",
    "endpoint": "/chat/completions",
    "errors": null,
    "input_file_id": "file-b4fa7277f3eb4722ac18b90f46102c3f",
    "completion_window": "24h",
    "status": "cancelling",
    "output_file_id": null,
    "error_file_id": null,
    "created_at": 1646126127,
    "in_progress_at": 1646126130,
    "expires_at": 1646126170,
    "finalizing_at": null,
    "completed_at": null,
    "failed_at": null,
    "expired_at": null,
    "cancelling_at": 1646126138,
    "cancelled_at": null,
    "request_counts": {
      "total": 500,
      "completed": 100,
      "failed": 5
    },
    "metadata": {
      "batch_description": "Evaluation job"
    }
  }
}

저장 완성 - 목록

GET {endpoint}/openai/chat/completions?api-version=2025-04-01-preview

저장된 완성 목록이 나옵니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
metadata query No string 저장된 완성 함수의 (키, 값) 쌍으로 필터링합니다.
model query No string 모델명으로 필터링하세요.
after query No string 이전 페이지네이션 요청에서 마지막으로 저장된 완료 식별자.
limit query No integer 복구해야 할 저장된 완성 파일 수. 기본값은 20입니다.
order query No string 결과의 생성 시간(오름 또는 하강)에 따른 순서. 기본적으로 설명으로 설정하세요.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json StoredCompletionList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/chat/completions?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "list",
    "data": [
      {
        "id": "chatcmpl-AvBCKqg2xqxVHCCEnUB4Bgj1Bjl7Y",
        "model": "gpt-4o-mini-2024-07-18",
        "created": 1738193475,
        "request_id": "e539c139-a97d-4ee1-bde9-3be3d5c6edb5",
        "usage": {
          "total_tokens": 25,
          "completion_tokens": 7,
          "prompt_tokens": 18
        },
        "seed": 123,
        "top_p": 1,
        "temperature": 1,
        "presence_penalty": 0,
        "frequency_penalty": 0,
        "system_fingerprint": "fp_f3927aa00d",
        "metadata": {
          "key_1": "val_1",
          "key_2": "val_2"
        },
        "choices": [
          {
            "index": 0,
            "message": {
              "content": "Hello, how are you?",
              "role": "assistant"
            },
            "finish_reason": "stop"
          }
        ]
      }
    ],
    "total": 1,
    "first_id": "chatcmpl-AvBCKqg2xqxVHCCEnUB4Bgj1Bjl7Y",
    "last_id": "chatcmpl-AvBCKqg2xqxVHCCEnUB4Bgj1Bjl7Y",
    "has_more": false
  }
}

저장된 완성 - Get

GET {endpoint}/openai/chat/completions/{stored-completion-id}?api-version=2025-04-01-preview

저장된 완료 ID에 의해 저장된 완성 상태를 받습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
stored-completion-id path Yes string 저장된 완료의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json StoredCompletionResponse

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/chat/completions/{stored-completion-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "chat.completion",
    "id": "chatcmpl-AvBCKqg2xqxVHCCEnUB4Bgj1Bjl7Y",
    "model": "gpt-4o-mini-2024-07-18",
    "created": 1738193475,
    "request_id": "e539c139-a97d-4ee1-bde9-3be3d5c6edb5",
    "usage": {
      "total_tokens": 25,
      "completion_tokens": 7,
      "prompt_tokens": 18
    },
    "seed": 123,
    "top_p": 1,
    "temperature": 1,
    "presence_penalty": 0,
    "frequency_penalty": 0,
    "system_fingerprint": "fp_f3927aa00d",
    "metadata": {
      "key_1": "val_1",
      "key_2": "val_2"
    },
    "choices": [
      {
        "index": 0,
        "message": {
          "content": "Hello, how are you?",
          "role": "assistant"
        },
        "finish_reason": "stop"
      }
    ]
  }
}

저장된 완성 - 업데이트

POST {endpoint}/openai/chat/completions/{stored-completion-id}?api-version=2025-04-01-preview

저장된 완료를 주어진 저장 완료 ID로 업데이트합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
stored-completion-id path Yes string 저장된 완료의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: application/json

Name Type Description Required Default
metadata object 추가 정보를 위한 임의의 키-값 쌍. No

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json StoredCompletion

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/chat/completions/{stored-completion-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "chat.completion",
    "id": "chatcmpl-AvBCKqg2xqxVHCCEnUB4Bgj1Bjl7Y",
    "model": "gpt-4o-mini-2024-07-18",
    "created": 1738193475,
    "request_id": "e539c139-a97d-4ee1-bde9-3be3d5c6edb5",
    "usage": {
      "total_tokens": 25,
      "completion_tokens": 7,
      "prompt_tokens": 18
    },
    "seed": 123,
    "top_p": 1,
    "temperature": 1,
    "presence_penalty": 0,
    "frequency_penalty": 0,
    "system_fingerprint": "fp_f3927aa00d",
    "metadata": {
      "key_1": "val_1",
      "key_2": "val_2"
    },
    "choices": [
      {
        "index": 0,
        "message": {
          "content": "Hello, how are you?",
          "role": "assistant"
        },
        "finish_reason": "stop"
      }
    ]
  }
}

저장된 완성 - 삭제

DELETE {endpoint}/openai/chat/completions/{stored-completion-id}?api-version=2025-04-01-preview

저장된 완료 ID를 저장한 완료 ID로 삭제합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
stored-completion-id path Yes string 저장된 완료의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 저장된 완료 완료가 성공적으로 삭제되었습니다.

Content-Type Type Description
application/json StoredCompletionDelete

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

DELETE {endpoint}/openai/chat/completions/{stored-completion-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "chat.completion.deleted",
    "id": "chatcmpl-AvBCKqg2xqxVHCCEnUB4Bgj1Bjl7Y",
    "deleted": true
  }
}

저장된 완성 - Getmessages

GET {endpoint}/openai/chat/completions/{stored-completion-id}/messages?api-version=2025-04-01-preview

주어진 저장된 완료 ID에 의해 저장된 완료 메시지를 받습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
stored-completion-id path Yes string 저장된 완료의 식별자입니다.
after query No string 이전 페이지네이션 요청에서 마지막으로 저장된 완료 메시지의 식별자입니다.
limit query No integer 복구해야 할 저장된 완성 메시지 수. 기본값은 20입니다.
order query No string 결과 순서는 메시지 인덱스(상승 또는 하강)에 따라 정렬됩니다. 기본적으로 상승(ASC)으로 설정하세요.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json StoredCompletionMessages

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/chat/completions/{stored-completion-id}/messages?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "list",
    "data": [
      {
        "role": "user",
        "content": "Hello"
      }
    ],
    "total": 1,
    "first_id": "chatcmpl-AvBCKqg2xqxVHCCEnUB4Bgj1Bjl7Y",
    "last_id": "chatcmpl-AvBCKqg2xqxVHCCEnUB4Bgj1Bjl7Y",
    "has_more": false
  }
}

평가 - Get 목록

GET {endpoint}/openai/evals?api-version=2025-04-01-preview

프로젝트 평가 목록을 작성하세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
after query No string 이전 페이지네이션 요청에서 나온 마지막 평가 식별자입니다.
limit query No integer 복구해야 할 평가 수.
order query No string
가능한 값들: asc, desc
평가 순서를 타임스탬프별로 정렬하세요. 오름차순 또는 desc 내림순으로 사용 asc 하세요.
order_by query No string
가능한 값들: created_at, updated_at
평가는 생성 시간 또는 마지막 업데이트 시간 순으로 정렬할 수 있습니다. 생성 시간 또는 updated_at 마지막 업데이트 시간 표시로 사용 created_at 하세요.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 평가 목록

Content-Type Type Description
application/json EvalList

Examples

GET {endpoint}/openai/evals?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "list",
    "data": [
      {
        "object": "eval",
        "id": "eval_6801694950848190b10968bb628b651d",
        "data_source_config": {
          "type": "custom",
          "schema": {
            "type": "object",
            "properties": {
              "item": {
                "type": "object",
                "properties": {
                  "question": {
                    "type": "string"
                  },
                  "A": {
                    "type": "string"
                  },
                  "B": {
                    "type": "string"
                  },
                  "C": {
                    "type": "string"
                  },
                  "D": {
                    "type": "string"
                  },
                  "answer": {
                    "type": "string"
                  }
                }
              }
            },
            "required": [
              "item"
            ]
          }
        },
        "testing_criteria": [
          {
            "name": "string check",
            "type": "string_check",
            "input": "{{sample.output_text}}",
            "reference": "{{item.answer}}",
            "operation": "eq"
          }
        ],
        "name": "Math Quiz",
        "created_at": 1744922953,
        "metadata": {}
      }
    ],
    "first_id": "eval_6801694950848190b10968bb628b651d",
    "last_id": "eval_6801694950848190b10968bb628b651d",
    "has_more": false
  }
}

평가 - 창조

POST {endpoint}/openai/evals?api-version=2025-04-01-preview

모델의 성능을 테스트하는 데 사용할 수 있는 평가 구조를 만드세요. 평가는 일련의 테스트 기준과 데이터소스입니다. 평가를 생성한 후에는 다양한 모델과 모델 매개변수에 대해 실행할 수 있습니다. 우리는 여러 유형의 채점기와 데이터소스를 지원합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 201

Description: OK

Content-Type Type Description
application/json Eval

Examples

POST {endpoint}/openai/evals?api-version=2025-04-01-preview

{
 "name": "Math Quiz",
 "data_source_config": {
  "type": "custom",
  "include_sample_schema": true,
  "item_schema": {
   "type": "object",
   "properties": {
    "question": {
     "type": "string"
    },
    "A": {
     "type": "string"
    },
    "B": {
     "type": "string"
    },
    "C": {
     "type": "string"
    },
    "D": {
     "type": "string"
    },
    "answer": {
     "type": "string"
    }
   }
  }
 },
 "testing_criteria": [
  {
   "type": "string_check",
   "reference": "{{item.answer}}",
   "input": "{{sample.output_text}}",
   "operation": "eq",
   "name": "string check"
  }
 ]
}

응답: 상태 코드: 201

{
  "headers": {},
  "body": {
    "object": "eval",
    "id": "eval_6801694950848190b10968bb628b651d",
    "data_source_config": {
      "type": "custom",
      "schema": {
        "type": "object",
        "properties": {
          "item": {
            "type": "object",
            "properties": {
              "question": {
                "type": "string"
              },
              "A": {
                "type": "string"
              },
              "B": {
                "type": "string"
              },
              "C": {
                "type": "string"
              },
              "D": {
                "type": "string"
              },
              "answer": {
                "type": "string"
              }
            }
          }
        },
        "required": [
          "item"
        ]
      }
    },
    "testing_criteria": [
      {
        "name": "string check",
        "type": "string_check",
        "input": "{{sample.output_text}}",
        "reference": "{{item.answer}}",
        "operation": "eq"
      }
    ],
    "name": "Math Quiz",
    "created_at": 1744922953,
    "metadata": {}
  }
}

평가 - 삭제

DELETE {endpoint}/openai/evals/{eval-id}?api-version=2025-04-01-preview

평가를 삭제하세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 삭제할 평가 ID입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 평가를 성공적으로 삭제했습니다.

Content-Type Type Description
application/json object

상태 코드: 404

설명: 평가 없음.

Content-Type Type Description
application/json Error

Examples

DELETE {endpoint}/openai/evals/{eval-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "eval.deleted",
    "deleted": true,
    "eval_id": "eval_6801694950848190b10968bb628b651d"
  }
}

상태 코드: 404

{
  "headers": {},
  "body": {
    "code": "notFound",
    "message": "Evaluation with ID eval_6801694950848190b10968bb628b651d not found."
  }
}

평가 - Get

GET {endpoint}/openai/evals/{eval-id}?api-version=2025-04-01-preview

신분증으로 평가를 받으세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 평가 ID를 찾아야 합니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 평가

Content-Type Type Description
application/json Eval

Examples

GET {endpoint}/openai/evals/{eval-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "eval",
    "id": "eval_6801694950848190b10968bb628b651d",
    "data_source_config": {
      "type": "custom",
      "schema": {
        "type": "object",
        "properties": {
          "item": {
            "type": "object",
            "properties": {
              "question": {
                "type": "string"
              },
              "A": {
                "type": "string"
              },
              "B": {
                "type": "string"
              },
              "C": {
                "type": "string"
              },
              "D": {
                "type": "string"
              },
              "answer": {
                "type": "string"
              }
            }
          }
        },
        "required": [
          "item"
        ]
      }
    },
    "testing_criteria": [
      {
        "name": "string check",
        "type": "string_check",
        "input": "{{sample.output_text}}",
        "reference": "{{item.answer}}",
        "operation": "eq"
      }
    ],
    "name": "Math Quiz",
    "created_at": 1744922953,
    "metadata": {}
  }
}

평가 - 업데이트

POST {endpoint}/openai/evals/{eval-id}?api-version=2025-04-01-preview

평가의 특정 속성을 업데이트하세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 평가 ID를 업데이트하세요.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 업데이트된 평가

Content-Type Type Description
application/json Eval

Examples

POST {endpoint}/openai/evals/{eval-id}?api-version=2025-04-01-preview

{
 "name": "Updated Math Quiz",
 "metadata": {
  "description": "Updated description"
 }
}

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "eval",
    "id": "eval_6801694950848190b10968bb628b651d",
    "data_source_config": {
      "type": "custom",
      "schema": {
        "type": "object",
        "properties": {
          "item": {
            "type": "object",
            "properties": {
              "question": {
                "type": "string"
              },
              "A": {
                "type": "string"
              },
              "B": {
                "type": "string"
              },
              "C": {
                "type": "string"
              },
              "D": {
                "type": "string"
              },
              "answer": {
                "type": "string"
              }
            }
          }
        },
        "required": [
          "item"
        ]
      }
    },
    "testing_criteria": [
      {
        "name": "string check",
        "type": "string_check",
        "input": "{{sample.output_text}}",
        "reference": "{{item.answer}}",
        "operation": "eq"
      }
    ],
    "name": "Updated Math Quiz",
    "created_at": 1744922953,
    "metadata": {
      "description": "Updated description"
    }
  }
}

평가 - Getrunlist

GET {endpoint}/openai/evals/{eval-id}/runs?api-version=2025-04-01-preview

평가를 위해 달리기 목록을 받으세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 검색할 평가의 ID는 적용됩니다.
after query No string 이전 페이지네이션 요청에서 마지막 실행을 위한 식별자입니다.
limit query No integer 회수해야 할 득점 수.
order query No string
가능한 값들: asc, desc
런 순서를 타임스탬프별로 정렬하세요. 오름차순 또는 desc 내림순으로 사용 asc 하세요. 기본값은 asc입니다.
status query No string
가능한 값: queued, in_progress, completed, canceled, , failed
필터는 상태별로 작동합니다. "대기", "in_progress", "실패", "완료", "취소"를 사용하세요.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 평가를 위한 런 목록

Content-Type Type Description
application/json EvalRunList

Examples

GET {endpoint}/openai/evals/{eval-id}/runs?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "list",
    "data": [
      {
        "object": "eval.run",
        "id": "evalrun_68016a056f0481909b9774447bdd1aa3",
        "eval_id": "eval_6801694950848190b10968bb628b651d",
        "report_url": "https://ai.azure.com/resource/evaluation",
        "status": "queued",
        "model": "gpt-4o-mini",
        "name": "Math quiz",
        "created_at": 1744923141,
        "result_counts": {
          "total": 0,
          "errored": 0,
          "failed": 0,
          "passed": 0
        },
        "per_model_usage": null,
        "per_testing_criteria_results": null,
        "data_source": {
          "type": "completions",
          "source": {
            "type": "file_content",
            "content": [
              {
                "item": {
                  "question": "Find the degree for the given field extension Q(sqrt(2), sqrt(3), sqrt(18)) over Q.",
                  "A": "0",
                  "B": "4",
                  "C": "2",
                  "D": "6",
                  "answer": "B"
                }
              },
              {
                "item": {
                  "question": "Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5.",
                  "A": "8",
                  "B": "2",
                  "C": "24",
                  "D": "120",
                  "answer": "C"
                }
              }
            ]
          },
          "input_messages": {
            "template": [
              {
                "type": "message",
                "role": "system",
                "content": {
                  "text": "Answer the question's with A, B, C, or D."
                }
              },
              {
                "type": "message",
                "role": "user",
                "content": {
                  "text": "Question: {{item.question}} A: {{item.A}} B: {{item.B}} C: {{item.C}} D: {{item.D}}."
                }
              }
            ]
          },
          "model": "gpt-4o-mini",
          "sampling_params": {
            "seed": 42,
            "temperature": 1,
            "top_p": 1
          }
        },
        "error": null,
        "metadata": {}
      }
    ],
    "first_id": "evalrun_68016a056f0481909b9774447bdd1aa3",
    "last_id": "evalrun_68016a056f0481909b9774447bdd1aa3",
    "has_more": false
  }
}

평가 - 생성 실행

POST {endpoint}/openai/evals/{eval-id}/runs?api-version=2025-04-01-preview

새로운 평가 실행을 생성하세요. 이것이 채점의 시작점입니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 런 포를 만들기 위한 평가의 ID입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 201

설명: 평가를 위한 런을 성공적으로 생성함

Content-Type Type Description
application/json EvalRun

상태 코드: 400

설명: 나쁜 요청(예: 평가 대상 누락)

Content-Type Type Description
application/json Error

Examples

POST {endpoint}/openai/evals/{eval-id}/runs?api-version=2025-04-01-preview

{
 "name": "Math quiz",
 "data_source": {
  "type": "completions",
  "source": {
   "type": "file_content",
   "content": [
    {
     "item": {
      "question": "Find the degree for the given field extension Q(sqrt(2), sqrt(3), sqrt(18)) over Q.",
      "A": "0",
      "B": "4",
      "C": "2",
      "D": "6",
      "answer": "B"
     }
    },
    {
     "item": {
      "question": "Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5.",
      "A": "8",
      "B": "2",
      "C": "24",
      "D": "120",
      "answer": "C"
     }
    }
   ]
  },
  "input_messages": {
   "type": "template",
   "template": [
    {
     "type": "message",
     "role": "system",
     "content": {
      "text": "Answer the question's with A, B, C, or D."
     }
    },
    {
     "type": "message",
     "role": "user",
     "content": {
      "text": "Question: {{item.question}} A: {{item.A}} B: {{item.B}} C: {{item.C}} D: {{item.D}}."
     }
    }
   ]
  },
  "model": "gpt-4o-mini",
  "sampling_params": {
   "temperature": 1,
   "top_p": 1,
   "seed": 42
  }
 }
}

응답: 상태 코드: 201

{
  "headers": {},
  "body": {
    "object": "eval.run",
    "id": "evalrun_68016a056f0481909b9774447bdd1aa3",
    "eval_id": "eval_6801694950848190b10968bb628b651d",
    "report_url": "https://ai.azure.com/resource/evaluation",
    "status": "queued",
    "model": "gpt-4o-mini",
    "name": "Math quiz",
    "created_at": 1744923141,
    "result_counts": {
      "total": 0,
      "errored": 0,
      "failed": 0,
      "passed": 0
    },
    "per_model_usage": null,
    "per_testing_criteria_results": null,
    "data_source": {
      "type": "completions",
      "source": {
        "type": "file_content",
        "content": [
          {
            "item": {
              "question": "Find the degree for the given field extension Q(sqrt(2), sqrt(3), sqrt(18)) over Q.",
              "A": "0",
              "B": "4",
              "C": "2",
              "D": "6",
              "answer": "B"
            }
          },
          {
            "item": {
              "question": "Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5.",
              "A": "8",
              "B": "2",
              "C": "24",
              "D": "120",
              "answer": "C"
            }
          }
        ]
      },
      "input_messages": {
        "type": "template",
        "template": [
          {
            "type": "message",
            "role": "system",
            "content": {
              "text": "Answer the question's with A, B, C, or D."
            }
          },
          {
            "type": "message",
            "role": "user",
            "content": {
              "text": "Question: {{item.question}} A: {{item.A}} B: {{item.B}} C: {{item.C}} D: {{item.D}}."
            }
          }
        ]
      },
      "model": "gpt-4o-mini",
      "sampling_params": {
        "seed": 42,
        "temperature": 1,
        "top_p": 1
      }
    },
    "error": null,
    "metadata": {}
  }
}

평가 - 삭제 실행

DELETE {endpoint}/openai/evals/{eval-id}/runs/{run-id}?api-version=2025-04-01-preview

평가 실행 하나를 삭제하세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 런을 삭제할 평가의 ID입니다.
run-id path Yes string 삭제 실행 ID입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 평가 실행 파일을 성공적으로 삭제함

Content-Type Type Description
application/json object

상태 코드: 404

설명: 런을 찾지 못했습니다

Content-Type Type Description
application/json Error

Examples

DELETE {endpoint}/openai/evals/{eval-id}/runs/{run-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "eval.deleted",
    "deleted": true,
    "run_id": "evalrun_68016a056f0481909b9774447bdd1aa3"
  }
}

상태 코드: 404

{
  "headers": {},
  "body": {
    "code": "notFound",
    "message": "Evaluation Run with ID evalrun_68016a056f0481909b9774447bdd1aa3 not found."
  }
}

평가 - 런

GET {endpoint}/openai/evals/{eval-id}/runs/{run-id}?api-version=2025-04-01-preview

ID에서 평가를 받으세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 검색할 평가의 ID는 적용됩니다.
run-id path Yes string 회수할 수 있는 런의 ID입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 평가 실행

Content-Type Type Description
application/json EvalRun

Examples

GET {endpoint}/openai/evals/{eval-id}/runs/{run-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "eval.run",
    "id": "evalrun_68016a056f0481909b9774447bdd1aa3",
    "eval_id": "eval_6801694950848190b10968bb628b651d",
    "report_url": "https://ai.azure.com/resource/evaluation",
    "status": "queued",
    "model": "gpt-4o-mini",
    "name": "Math quiz",
    "created_at": 1744923141,
    "result_counts": {
      "total": 0,
      "errored": 0,
      "failed": 0,
      "passed": 0
    },
    "per_model_usage": null,
    "per_testing_criteria_results": null,
    "data_source": {
      "type": "completions",
      "source": {
        "type": "file_content",
        "content": [
          {
            "item": {
              "question": "Find the degree for the given field extension Q(sqrt(2), sqrt(3), sqrt(18)) over Q.",
              "A": "0",
              "B": "4",
              "C": "2",
              "D": "6",
              "answer": "B"
            }
          },
          {
            "item": {
              "question": "Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5.",
              "A": "8",
              "B": "2",
              "C": "24",
              "D": "120",
              "answer": "C"
            }
          }
        ]
      },
      "input_messages": {
        "type": "template",
        "template": [
          {
            "type": "message",
            "role": "system",
            "content": {
              "text": "Answer the question's with A, B, C, or D."
            }
          },
          {
            "type": "message",
            "role": "user",
            "content": {
              "text": "Question: {{item.question}} A: {{item.A}} B: {{item.B}} C: {{item.C}} D: {{item.D}}."
            }
          }
        ]
      },
      "model": "gpt-4o-mini",
      "sampling_params": {
        "seed": 42,
        "temperature": 1,
        "top_p": 1
      }
    },
    "error": null,
    "metadata": {}
  }
}

평가 - 실행 취소

POST {endpoint}/openai/evals/{eval-id}/runs/{run-id}?api-version=2025-04-01-preview

진행 중인 평가 작업을 취소하세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 취소하고 싶은 평가의 ID입니다.
run-id path Yes string 취소 신청 ID입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 취소된 평가 실행 객체

Content-Type Type Description
application/json EvalRun

Examples

POST {endpoint}/openai/evals/{eval-id}/runs/{run-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "eval.run",
    "id": "evalrun_68016a056f0481909b9774447bdd1aa3",
    "eval_id": "eval_6801694950848190b10968bb628b651d",
    "report_url": "https://ai.azure.com/resource/evaluation",
    "status": "canceled",
    "model": "gpt-4o-mini",
    "name": "Math quiz",
    "created_at": 1744923141,
    "result_counts": {
      "total": 0,
      "errored": 0,
      "failed": 0,
      "passed": 0
    },
    "per_model_usage": null,
    "per_testing_criteria_results": null,
    "data_source": {
      "type": "completions",
      "source": {
        "type": "file_content",
        "content": [
          {
            "item": {
              "question": "Find the degree for the given field extension Q(sqrt(2), sqrt(3), sqrt(18)) over Q.",
              "A": "0",
              "B": "4",
              "C": "2",
              "D": "6",
              "answer": "B"
            }
          },
          {
            "item": {
              "question": "Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5.",
              "A": "8",
              "B": "2",
              "C": "24",
              "D": "120",
              "answer": "C"
            }
          }
        ]
      },
      "input_messages": {
        "type": "template",
        "template": [
          {
            "type": "message",
            "role": "system",
            "content": {
              "text": "Answer the question's with A, B, C, or D."
            }
          },
          {
            "type": "message",
            "role": "user",
            "content": {
              "text": "Question: {{item.question}} A: {{item.A}} B: {{item.B}} C: {{item.C}} D: {{item.D}}."
            }
          }
        ]
      },
      "model": "gpt-4o-mini",
      "sampling_params": {
        "seed": 42,
        "temperature": 1,
        "top_p": 1
      }
    },
    "error": null,
    "metadata": {}
  }
}

평가 - Getrunoutputitems

GET {endpoint}/openai/evals/{eval-id}/runs/{run-id}/output_items?api-version=2025-04-01-preview

평가 실행을 위한 출력 항목 목록을 받으세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 검색할 평가의 ID는 적용됩니다.
run-id path Yes string 출력 항목을 가져오기 위한 실행 ID입니다.
after query No string 이전 페이지네이션 요청에서 마지막으로 출력된 항목에 대한 식별자.
limit query No integer 회수해야 할 출력 항목 수.
status query No string
가능한 값들: fail, pass
출력 항목은 상태별로 필터링하세요. 실패한 출력 항목별로 필터링하거나 pass 패스된 출력 항목별로 필터링하는 데 사용 failed 하세요.
order query No string
가능한 값들: asc, desc
출력 항목의 순서를 타임스탬프별로 정렬하세요. 오름차순 또는 desc 내림순으로 사용 asc 하세요. 기본값은 asc입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 평가 실행을 위한 출력 항목 목록

Content-Type Type Description
application/json EvalRunOutputItemList

Examples

GET {endpoint}/openai/evals/{eval-id}/runs/{run-id}/output_items?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "list",
    "data": [
      {
        "object": "eval.run.output_item",
        "id": "outputitem_68017251ff3881908bf5096bf4cd91c6",
        "created_at": 1744925265,
        "run_id": "evalrun_68016a056f0481909b9774447bdd1aa3",
        "eval_id": "eval_6801694950848190b10968bb628b651d",
        "status": "fail",
        "datasource_item_id": 1,
        "datasource_item": {
          "question": "Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5.",
          "A": "8",
          "B": "2",
          "C": "24",
          "D": "120",
          "answer": "C"
        },
        "results": [
          {
            "name": "string check-63b1fffa-bee6-4c37-ae77-ed46e3dce2b7",
            "sample": null,
            "passed": false,
            "score": 0
          }
        ],
        "sample": {
          "input": [
            {
              "role": "system",
              "content": "Answer the question's with A, B, C, or D.",
              "function_call": null
            },
            {
              "role": "user",
              "content": "Question: Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5. A: 8 B: 2 C: 24 D: 120.",
              "function_call": null
            }
          ],
          "output": [
            {
              "role": "assistant",
              "content": "To find the index of the subgroup generated by the permutation \\( p = (1, 2, 5, 4)(2, 3) \\) in \\( S_5 \\), we first need to determine the order of \\( p \\).\n\n1. The cycle \\( (1, 2, 5, 4) \\) has length 4.\n2. The cycle \\( (2, 3) \\) has length 2.\n3. The least common multiple (LCM) of these lengths gives the order of the permutation \\( p \\).\n\nCalculating the LCM:\n\\[\n\\text{lcm}(4, 2) = 4\n\\]\nThus, the order of \\( p \\) is 4.\n\nNext, we find the size of the subgroup \\( \\langle p \\rangle \\):\n- The size of \\( \\langle p \\rangle \\) is equal to the order of \\( p \\), which is 4.\n\nNow, we know the size of \\( S_5 \\):\n\\[\n|S_5| = 5! = 120\n\\]\n\nFinally, to find the index of \\( \\langle p \\rangle \\) in \\( S_5 \\), we use the formula:\n\\[\n\\text{index} = \\frac{|S_5|}{|\\langle p \\rangle|} = \\frac{120}{4} = 30\n\\]\n\nHowever, the available answer choices do not include 30. There may be a mistake because the question expects an answer among A, B, C, and D.\n\nLet\u00e2\u20ac\u2122s assume we misunderstood \\( \\langle p \\rangle \\), and instead, we can deduce based solely on given answers looking for a relation to \\( S_5 \\) without delving into detailed subgroup tracks. \n\nSince \\( S_5 \\) has 30 different elements in a subgroup configuration, the closest answer physically relating as long as \\( p \\) covers two elements effectively would logically fit an answer of 120 / 60 which has no direct relationship.\n\nGiven the option choices and specific rank formulations as often made regarding elements in specific construct the subgroup at best reflects around a viable ratio of parts allowed through available indices. \n\nThus, after reasoning through which aligns most structurally geometrically yielding across all configurations possible integrated yet arrives leading \\( p \\) through neighborhood distributions leaving reflections outstanding:\n\n\n\nThe correct answer is:\n**C: 24**\n\nHowever per the discussion migrating \\( p \\) may leave various pathways leading ultimately toward that framing in modeling. Always a good suggestion confirming the elements group generating any possible in outputs rationales. ",
              "function_call": null
            }
          ],
          "finish_reason": "stop",
          "model": "gpt-4o-mini-2024-07-18",
          "usage": {
            "total_tokens": 627,
            "completion_tokens": 546,
            "prompt_tokens": 81,
            "cached_tokens": 0
          },
          "error": null,
          "temperature": 1,
          "top_p": 1,
          "seed": 42
        }
      },
      {
        "object": "eval.run.output_item",
        "id": "outputitem_6801724f54888190942ec66a197309dd",
        "created_at": 1744925263,
        "run_id": "evalrun_68016a056f0481909b9774447bdd1aa3",
        "eval_id": "eval_6801694950848190b10968bb628b651d",
        "status": "fail",
        "datasource_item_id": 0,
        "datasource_item": {
          "question": "Find the degree for the given field extension Q(sqrt(2), sqrt(3), sqrt(18)) over Q.",
          "A": "0",
          "B": "4",
          "C": "2",
          "D": "6",
          "answer": "B"
        },
        "results": [
          {
            "name": "string check-63b1fffa-bee6-4c37-ae77-ed46e3dce2b7",
            "sample": null,
            "passed": false,
            "score": 0
          }
        ],
        "sample": {
          "input": [
            {
              "role": "system",
              "content": "Answer the question's with A, B, C, or D.",
              "function_call": null
            },
            {
              "role": "user",
              "content": "Question: Find the degree for the given field extension Q(sqrt(2), sqrt(3), sqrt(18)) over Q. A: 0 B: 4 C: 2 D: 6.",
              "function_call": null
            }
          ],
          "output": [
            {
              "role": "assistant",
              "content": "To find the degree of the field extension \\(\\mathbb{Q}(\\sqrt{2}, \\sqrt{3}, \\sqrt{18})\\) over \\(\\mathbb{Q}\\), we can simplify \\(\\sqrt{18}\\) since \\(\\sqrt{18} = \\sqrt{9 \\cdot 2} = 3\\sqrt{2}\\). \n\nThus, the field can be rewritten as \\(\\mathbb{Q}(\\sqrt{2}, \\sqrt{3})\\).\n\nNext, we calculate the degree of the extensions step by step:\n\n1. The extension \\(\\mathbb{Q}(\\sqrt{2})\\) over \\(\\mathbb{Q}\\) has degree 2, since \\(\\sqrt{2}\\) is not in \\(\\mathbb{Q}\\) and satisfies the polynomial \\(x^2 - 2\\).\n\n2. Next, we consider the extension \\(\\mathbb{Q}(\\sqrt{2}, \\sqrt{3})\\) over \\(\\mathbb{Q}(\\sqrt{2})\\). The element \\(\\sqrt{3}\\) is also not in \\(\\mathbb{Q}(\\sqrt{2})\\) and satisfies the polynomial \\(x^2 - 3\\), which is irreducible over \\(\\mathbb{Q}(\\sqrt{2})\\). Hence, the degree of the extension \\(\\mathbb{Q}(\\sqrt{2}, \\sqrt{3})\\) over \\(\\mathbb{Q}(\\sqrt{2})\\) is also 2.\n\nNow we can combine these degrees:\n\n\\[\n[\\mathbb{Q}(\\sqrt{2}, \\sqrt{3}) : \\mathbb{Q}] = [\\mathbb{Q}(\\sqrt{2}, \\sqrt{3}) : \\mathbb{Q}(\\sqrt{2})] \\times [\\mathbb{Q}(\\sqrt{2}) : \\mathbb{Q}] = 2 \\times 2 = 4.\n\\]\n\nThus, the degree of the field extension \\(\\mathbb{Q}(\\sqrt{2}, \\sqrt{3}, \\sqrt{18})\\) over \\(\\mathbb{Q}\\) is 4.\n\nTherefore, the answer is:\n\n**B: 4**.",
              "function_call": null
            }
          ],
          "finish_reason": "stop",
          "model": "gpt-4o-mini-2024-07-18",
          "usage": {
            "total_tokens": 556,
            "completion_tokens": 487,
            "prompt_tokens": 69,
            "cached_tokens": 0
          },
          "error": null,
          "temperature": 1,
          "top_p": 1,
          "seed": 42
        }
      }
    ],
    "first_id": "outputitem_68017251ff3881908bf5096bf4cd91c6",
    "last_id": "outputitem_6801724f54888190942ec66a197309dd",
    "has_more": false
  }
}

평가 - Getrunoutputitem

GET {endpoint}/openai/evals/{eval-id}/runs/{run-id}/output_items/{output-item-id}?api-version=2025-04-01-preview

ID별로 평가 실행 결과물을 얻으세요.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
eval-id path Yes string 검색할 평가의 ID는 적용됩니다.
run-id path Yes string 회수할 수 있는 런의 ID입니다.
output-item-id path Yes string 가져올 출력 항목의 ID입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 평가 실행 출력 항목

Content-Type Type Description
application/json EvalRunOutputItem

Examples

GET {endpoint}/openai/evals/{eval-id}/runs/{run-id}/output_items/{output-item-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "eval.run.output_item",
    "id": "outputitem_68017251ff3881908bf5096bf4cd91c6",
    "created_at": 1744925265,
    "run_id": "evalrun_68016a056f0481909b9774447bdd1aa3",
    "eval_id": "eval_6801694950848190b10968bb628b651d",
    "status": "fail",
    "datasource_item_id": 1,
    "datasource_item": {
      "question": "Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5.",
      "A": "8",
      "B": "2",
      "C": "24",
      "D": "120",
      "answer": "C"
    },
    "results": [
      {
        "name": "string check-63b1fffa-bee6-4c37-ae77-ed46e3dce2b7",
        "sample": null,
        "passed": false,
        "score": 0
      }
    ],
    "sample": {
      "input": [
        {
          "role": "system",
          "content": "Answer the question's with A, B, C, or D."
        },
        {
          "role": "user",
          "content": "Question: Let p = (1, 2, 5, 4)(2, 3) in S_5 . Find the index of <p> in S_5. A: 8 B: 2 C: 24 D: 120."
        }
      ],
      "output": [
        {
          "role": "assistant",
          "content": "To find the index of the subgroup generated by the permutation \\( p = (1, 2, 5, 4)(2, 3) \\) in \\( S_5 \\), we first need to determine the order of \\( p \\).\n\n1. The cycle \\( (1, 2, 5, 4) \\) has length 4.\n2. The cycle \\( (2, 3) \\) has length 2.\n3. The least common multiple (LCM) of these lengths gives the order of the permutation \\( p \\).\n\nCalculating the LCM:\n\\[\n\\text{lcm}(4, 2) = 4\n\\]\nThus, the order of \\( p \\) is 4.\n\nNext, we find the size of the subgroup \\( \\langle p \\rangle \\):\n- The size of \\( \\langle p \\rangle \\) is equal to the order of \\( p \\), which is 4.\n\nNow, we know the size of \\( S_5 \\):\n\\[\n|S_5| = 5! = 120\n\\]\n\nFinally, to find the index of \\( \\langle p \\rangle \\) in \\( S_5 \\), we use the formula:\n\\[\n\\text{index} = \\frac{|S_5|}{|\\langle p \\rangle|} = \\frac{120}{4} = 30\n\\]\n\nHowever, the available answer choices do not include 30. There may be a mistake because the question expects an answer among A, B, C, and D.\n\nLet\u00e2\u20ac\u2122s assume we misunderstood \\( \\langle p \\rangle \\), and instead, we can deduce based solely on given answers looking for a relation to \\( S_5 \\) without delving into detailed subgroup tracks. \n\nSince \\( S_5 \\) has 30 different elements in a subgroup configuration, the closest answer physically relating as long as \\( p \\) covers two elements effectively would logically fit an answer of 120 / 60 which has no direct relationship.\n\nGiven the option choices and specific rank formulations as often made regarding elements in specific construct the subgroup at best reflects around a viable ratio of parts allowed through available indices. \n\nThus, after reasoning through which aligns most structurally geometrically yielding across all configurations possible integrated yet arrives leading \\( p \\) through neighborhood distributions leaving reflections outstanding:\n\n\n\nThe correct answer is:\n**C: 24**\n\nHowever per the discussion migrating \\( p \\) may leave various pathways leading ultimately toward that framing in modeling. Always a good suggestion confirming the elements group generating any possible in outputs rationales. "
        }
      ],
      "finish_reason": "stop",
      "model": "gpt-4o-mini-2024-07-18",
      "usage": {
        "total_tokens": 627,
        "completion_tokens": 546,
        "prompt_tokens": 81,
        "cached_tokens": 0
      },
      "error": null,
      "temperature": 1,
      "top_p": 1,
      "seed": 42
    }
  }
}

파일 - 목록

GET {endpoint}/openai/files?api-version=2025-04-01-preview

Azure OpenAI 리소스가 소유한 모든 파일 목록을 얻습니다.

여기에는 미세 조정 모델의 훈련이나 검증을 위한 "파인튜닝" 파일과 같은 사용자가 업로드한 콘텐츠와, 해당 미세 조정 작업에 대한 다양한 지표를 포함하는 "파인-튜닝-결과"와 같이 서비스가 생성하는 파일이 포함됩니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
purpose query No string
가능한 값: fine-tune, fine-tune-results, assistants, assistants_output, batchbatch_output, , evals,evals_output
주어진 목적이 있는 파일만 반환하세요.
limit query No integer 반환할 물건 수 제한. 제한은 1에서 10,000까지 다양하며, 기본값은 10,000입니다.
after query No string 페이지네이션에 사용할 수 있는 커서입니다. after 이 객체 ID는 리스트 내 위치를 정의합니다. 예를 들어, 리스트 요청을 하고 obj_foo로 끝나는 100개의 객체를 받으면, 다음 호출에서 다음 페이지 가져오기를 위해 after=obj_foo를 포함할 수 있습니다.
order query No string 객체의 created_at 타임스탬프에 따라 순서를 정렬하세요. asc 오름차순과 desc 내림순에 대해서.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json FileList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/files?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "data": [
      {
        "status": "processed",
        "bytes": 140,
        "purpose": "fine-tune",
        "filename": "puppy.jsonl",
        "id": "file-181a1cbdcdcf4677ada87f63a0928099",
        "created_at": 1646126127,
        "object": "file"
      },
      {
        "status": "processed",
        "bytes": 32423,
        "purpose": "fine-tune-results",
        "filename": "results.csv",
        "id": "file-181a1cbdcdcf4677ada87f63a0928099",
        "created_at": 1646126127,
        "object": "file"
      }
    ],
    "object": "list"
  }
}

파일 - 업로드

POST {endpoint}/openai/files?api-version=2025-04-01-preview

로컬 머신에서 데이터를 업로드하여 새로운 파일 엔티티를 생성합니다. 업로드된 파일은 예를 들어, 미세 조정된 모델을 훈련하거나 평가하는 데 사용될 수 있습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: multipart/form-data

Name Type Description Required Default
anchor string 절대 만료 시점에 대해 앵커를 정의합니다. No
seconds integer 상대 유기량 시간을 몇 초 만에 알 수 있습니다.
범위: [1209600 - 2592000]. No
file string 파일을 받거나 설정하여 Azure OpenAI에 업로드합니다. Yes
purpose string 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes

Responses

상태 코드: 201

설명: 파일이 성공적으로 생성되었습니다.

Content-Type Type Description
application/json File

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/files?api-version=2025-04-01-preview

응답: 상태 코드: 201

{
  "headers": {
    "location": "https://aoairesource.openai.azure.com/openai/files/file-181a1cbdcdcf4677ada87f63a0928099"
  },
  "body": {
    "status": "pending",
    "purpose": "fine-tune",
    "filename": "puppy.jsonl",
    "id": "file-181a1cbdcdcf4677ada87f63a0928099",
    "created_at": 1646126127,
    "object": "file"
  }
}

파일 - 가져오기

POST {endpoint}/openai/files/import?api-version=2025-04-01-preview

제공된 URL에서 데이터를 가져와서 새로운 파일 엔티티를 생성합니다. 업로드된 파일은 예를 들어, 미세 조정된 모델을 훈련하거나 평가하는 데 사용될 수 있습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: application/json

Name Type Description Required Default
content_url string 문서를 다운로드할 URL은 (블롭의 SAS URL 또는 GET 요청으로 접근 가능한 다른 외부 URL 중 하나일 수 있습니다). Yes
expires_after FileExpiresAfter 파일의 만료일을 정의합니다. No
filename string 업로드할 JSON Lines 파일 이름입니다.
만약 ' purpose 파인튜닝'으로 설정되어 있다면, 각 줄은 'prompt'와 'completion' 필드가 포함된 JSON 레코드로 훈련 예제를 나타냅니다. Yes
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes

Responses

상태 코드: 201

설명: 파일이 성공적으로 생성되었습니다.

Content-Type Type Description
application/json File

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/files/import?api-version=2025-04-01-preview

응답: 상태 코드: 201

{
  "headers": {
    "location": "https://aoairesource.openai.azure.com/openai/files/file-181a1cbdcdcf4677ada87f63a0928099"
  },
  "body": {
    "status": "pending",
    "purpose": "fine-tune",
    "filename": "puppy.jsonl",
    "id": "file-181a1cbdcdcf4677ada87f63a0928099",
    "created_at": 1646126127,
    "object": "file"
  }
}

파일 - 받기

GET {endpoint}/openai/files/{file-id}?api-version=2025-04-01-preview

주어진 파일 ID로 지정된 단일 파일에 대해 상태, 크기, 목적 등 세부 정보를 제공합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
file-id path Yes string 파일의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json File

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/files/{file-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "status": "processed",
    "bytes": 140,
    "purpose": "fine-tune",
    "filename": "puppy.jsonl",
    "id": "file-181a1cbdcdcf4677ada87f63a0928099",
    "created_at": 1646126127,
    "object": "file"
  }
}

파일 - 삭제

DELETE {endpoint}/openai/files/{file-id}?api-version=2025-04-01-preview

주어진 파일 ID가 있는 파일을 삭제합니다.

파일이 미세 조정 작업에서 훈련 파일로 사용된 경우에도 삭제가 허용됩니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
file-id path Yes string 파일의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 파일이 성공적으로 삭제되었습니다.

Content-Type Type Description
application/json FileDelete

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

DELETE {endpoint}/openai/files/{file-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "object": "file",
    "deleted": true,
    "id": "file-181a1cbdcdcf4677ada87f63a0928099"
  }
}

파일 - 콘텐츠 받기

GET {endpoint}/openai/files/{file-id}/content?api-version=2025-04-01-preview

주어진 파일 ID로 지정된 파일의 내용을 가져옵니다.

파일은 사용자가 업로드한 콘텐츠일 수도 있고, 미세 조정 작업의 결과 지표처럼 서비스가 생성할 수도 있습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
file-id path Yes string 파일의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/octet-stream string
application/json string

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/octet-stream ErrorResponse
application/json ErrorResponse

Examples

GET {endpoint}/openai/files/{file-id}/content?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": "raw file content"
}

미세 조정 - 목록

GET {endpoint}/openai/fine_tuning/jobs?api-version=2025-04-01-preview

Azure OpenAI 리소스가 소유한 모든 미세 조정 작업 목록을 얻습니다. 각 미세 조정 작업에 반환되는 세부 정보에는 식별자 외에도 기본 모델, 훈련 및 검증 파일, 하이퍼 파라미터, 타임스탬프, 상태 및 이벤트가 포함됩니다.

직무 상태가 변경될 때 이벤트가 생성됩니다. 예를 들어, 실행 중이거나 완료 여부, 그리고 결과가 업로드된 시점입니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
after query No string 이전 페이지네이션 요청에서 발생한 마지막 이벤트의 식별자입니다.
limit query No integer 회수해야 할 이벤트 수. 기본값은 20입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json FineTuningJobList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/fine_tuning/jobs?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "has_more": false,
    "data": [
      {
        "hyperparameters": {
          "n_epochs": -1
        },
        "integrations": [
          {
            "type": "wandb",
            "wandb": {
              "project": "custom-wandb-project",
              "tags": [
                "project:tag",
                "lineage"
              ]
            }
          }
        ],
        "method": {
          "type": "supervised"
        },
        "status": "succeeded",
        "model": "curie",
        "fine_tuned_model": "curie.ft-72a2792ef7d24ba7b82c7fe4a37e379f",
        "training_file": "file-181a1cbdcdcf4677ada87f63a0928099",
        "result_files": [
          "file-181a1cbdcdcf4677ada87f63a0928099"
        ],
        "finished_at": 1646127311,
        "trained_tokens": 2342,
        "id": "ft-72a2792ef7d24ba7b82c7fe4a37e379f",
        "created_at": 1646126127,
        "object": "fine_tuning.job"
      }
    ],
    "object": "list"
  }
}

미세 조정 - 창조

POST {endpoint}/openai/fine_tuning/jobs?api-version=2025-04-01-preview

주어진 학습 파일에서 지정된 모델을 미세 조정하는 작업을 생성합니다.

응답에는 작업 상태와 하이퍼 매개변수 등 대기열에 저장된 작업의 세부 정보가 포함됩니다.

완성되면 미세 조정 모델의 이름이 응답에 추가됩니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: application/json

Name Type Description Required Default
hyperparameters FineTuningHyperParameters 미세 조정 작업에 사용되는 하이퍼 파라미터 설정들. No
integrations array 미세 조정 작업을 지원하는 통합 구성 목록입니다. 여러 통합 작업이 계획되어 있으니, 통합 유형을 확인하고 행동하는 것이 중요합니다. No
method object 감독 하 미세 조정에 사용되는 방법 No
└─ type FineTuningMethodType No
model string 이 미세 조정에 사용되는 기본 모델의 식별자(model-id)입니다. Yes
seed integer 미세 조정에 사용되는 씨앗. 동일한 시드와 작업 매개변수를 전달해도 결과는 동일하지만, 드물게 차이가 있을 수 있습니다. 시드가 지정되지 않았다면, 하나가 생성되어 줍니다. No
suffix string 미세 조정 모델을 식별하는 데 사용된 접미사. 접미사는 최대 40자(a-z, A-Z, 0-9, _)까지 포함할 수 있으며, 이는 미세 조정된 모델명에 추가됩니다. No
training_file string 이 미세 조정 모델을 학습하는 데 사용되는 파일 아이덴티티(file-id)입니다. Yes
validation_file string 훈련 중 미세 조정된 모델을 평가하는 데 사용되는 파일 식별자(file-id). No

Responses

상태 코드: 201

설명: 미세 조정이 성공적으로 제작되었습니다.

Content-Type Type Description
application/json FineTuningJob

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/fine_tuning/jobs?api-version=2025-04-01-preview

응답: 상태 코드: 201

{
  "headers": {
    "location": "https://aoairesource.openai.azure.com/openai/fine_tuning/jobs/ft-72a2792ef7d24ba7b82c7fe4a37e379f"
  },
  "body": {
    "hyperparameters": {
      "n_epochs": -1
    },
    "integrations": [
      {
        "type": "wandb",
        "wandb": {
          "project": "custom-wandb-project",
          "tags": [
            "project:tag",
            "lineage"
          ]
        }
      }
    ],
    "method": {
      "type": "supervised"
    },
    "status": "pending",
    "model": "curie",
    "training_file": "file-181a1cbdcdcf4677ada87f63a0928099",
    "id": "ft-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "fine_tuning.job"
  }
}

미세 조정 - 얻기

GET {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}?api-version=2025-04-01-preview

주어진 미세 조정 ID로 지정된 단일 미세 조정 작업에 대한 세부 정보를 얻습니다.

세부 정보에는 기본 모델, 훈련 및 검증 파일, 하이퍼 파라미터, 타임스탬프, 상태 및 이벤트가 포함됩니다.

이벤트는 작업 상태가 변경될 때(예: 실행 중이거나 완료됨), 그리고 결과가 업로드될 때 생성됩니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
fine-tuning-job-id path Yes string 미세 조정 작업의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json FineTuningJob

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "hyperparameters": {
      "n_epochs": -1
    },
    "integrations": [
      {
        "type": "wandb",
        "wandb": {
          "project": "custom-wandb-project",
          "tags": [
            "project:tag",
            "lineage"
          ]
        }
      }
    ],
    "method": {
      "type": "supervised"
    },
    "status": "succeeded",
    "model": "curie",
    "fine_tuned_model": "curie.ft-72a2792ef7d24ba7b82c7fe4a37e379f",
    "training_file": "file-181a1cbdcdcf4677ada87f63a0928099",
    "result_files": [
      "file-181a1cbdcdcf4677ada87f63a0928099"
    ],
    "finished_at": 1646127311,
    "trained_tokens": 2342,
    "id": "ft-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "fine_tuning.job"
  }
}

미세 조정 - 삭제

DELETE {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}?api-version=2025-04-01-preview

주어진 fine-tune-id에 의해 지정된 미세 조정 작업을 삭제합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
fine-tuning-job-id path Yes string 미세 조정 작업의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 204

설명: 미세 조정이 성공적으로 삭제되었습니다.

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

DELETE {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}?api-version=2025-04-01-preview

응답: 상태 코드: 204

{
  "headers": {}
}

미세 조정 - 취소

POST {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/cancel?api-version=2025-04-01-preview

주어진 미세 조정 ID에 의해 지정된 미세 조정 작업의 처리 과정을 취소합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
fine-tuning-job-id path Yes string 미세 조정 작업의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 미세 조정이 성공적으로 취소되었습니다

Content-Type Type Description
application/json FineTuningJob

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/cancel?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "hyperparameters": {
      "n_epochs": -1
    },
    "integrations": [
      {
        "type": "wandb",
        "wandb": {
          "project": "custom-wandb-project",
          "tags": [
            "project:tag",
            "lineage"
          ]
        }
      }
    ],
    "method": {
      "type": "supervised"
    },
    "status": "cancelled",
    "model": "curie",
    "training_file": "file-181a1cbdcdcf4677ada87f63a0928099",
    "finished_at": 1646127311,
    "id": "ft-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "fine_tuning.job"
  }
}

미세 조정 - 체크포인트 확보

GET {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/checkpoints?api-version=2025-04-01-preview

주어진 미세 조정 ID에 의해 지정된 미세 조정 작업에 대한 체크포인트를 얻습니다.

체크포인트는 훈련 중 성공적인 에포크가 끝날 때 생성됩니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
fine-tuning-job-id path Yes string 미세 조정 작업의 식별자입니다.
after query No string 이전 페이지네이션 요청에서 마지막 체크포인트 ID를 식별하는 식별자입니다.
limit query No integer 회수해야 할 체크포인트의 수. 기본값은 10입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json FineTuningJobCheckpointList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/checkpoints?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "has_more": false,
    "data": [
      {
        "id": "ftckpt_qtZ5Gyk4BLq1SfLFWp3RtO3P",
        "created_at": 1646126127,
        "fine_tuned_model_checkpoint": "gpt-35-turbo-0613.ft-72a2792ef7d24ba7b82c7fe4a37e379f:ckpt-step-10",
        "step_number": 10,
        "metrics": {
          "step": 10,
          "train_loss": 0.478,
          "train_mean_token_accuracy": 0.924,
          "valid_loss": 10.112,
          "valid_mean_token_accuracy": 0.145,
          "full_valid_loss": 0.567,
          "full_valid_mean_token_accuracy": 0.944
        },
        "fine_tuning_job_id": "ftjob-72a2792ef7d24ba7b82c7fe4a37e379f",
        "object": "fine_tuning.job.checkpoint"
      },
      {
        "id": "ftckpt_frtXDR2453C4FG67t03MLPs5",
        "created_at": 1646126137,
        "fine_tuned_model_checkpoint": "gpt-35-turbo-0613.ft-72a2792ef7d24ba7b82c7fe4a37e379f:ckpt-step-20",
        "step_number": 20,
        "metrics": {
          "step": 20,
          "train_loss": 0.355,
          "train_mean_token_accuracy": 0.947,
          "valid_loss": 11.32,
          "valid_mean_token_accuracy": 0.122,
          "full_valid_loss": 0.317,
          "full_valid_mean_token_accuracy": 0.955
        },
        "fine_tuning_job_id": "ftjob-72a2792ef7d24ba7b82c7fe4a37e379f",
        "object": "fine_tuning.job.checkpoint"
      },
      {
        "id": "ftckpt_agLk6Gio560ORp14gl123fgd",
        "created_at": 1646126147,
        "fine_tuned_model_checkpoint": "gpt-35-turbo-0613.ft-72a2792ef7d24ba7b82c7fe4a37e379f:ckpt-step-30",
        "step_number": 30,
        "metrics": {
          "step": 30,
          "train_loss": 0.155,
          "train_mean_token_accuracy": 0.975,
          "valid_loss": 9.31,
          "valid_mean_token_accuracy": 0.092,
          "full_valid_loss": 0.114,
          "full_valid_mean_token_accuracy": 0.963
        },
        "fine_tuning_job_id": "ftjob-72a2792ef7d24ba7b82c7fe4a37e379f",
        "object": "fine_tuning.job.checkpoint"
      }
    ],
    "object": "list"
  }
}

미세 조정 - 이벤트 획득

GET {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/events?api-version=2025-04-01-preview

주어진 미세 조정 ID에 의해 지정된 미세 조정 작업에 대한 이벤트를 받습니다.

이벤트는 작업 상태가 변경될 때(예: 실행 중이거나 완료됨), 그리고 결과가 업로드될 때 생성됩니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
fine-tuning-job-id path Yes string 미세 조정 작업의 식별자입니다.
after query No string 이전 페이지네이션 요청에서 발생한 마지막 이벤트의 식별자입니다.
limit query No integer 회수해야 할 이벤트 수. 기본값은 20입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json FineTuningJobEventList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/events?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "has_more": false,
    "data": [
      {
        "id": "ftevent-363dcd7cb4c74539bc53293c1dceef05",
        "created_at": 1646126127,
        "level": "info",
        "message": "Job enqueued. Waiting for jobs ahead to complete.",
        "type": "message",
        "object": "fine_tuning.job.event"
      },
      {
        "id": "ftevent-8c2a44999790437cb3230e543fa2cf0f",
        "created_at": 1646126169,
        "level": "info",
        "message": "Job started.",
        "type": "message",
        "object": "fine_tuning.job.event"
      },
      {
        "id": "ftevent-2d47d651d2f3484c8187c88c00078147",
        "created_at": 1646126192,
        "level": "info",
        "message": "Job succeeded.",
        "type": "message",
        "object": "fine_tuning.job.event"
      }
    ],
    "object": "list"
  }
}

미세 조정 - 일시정지

POST {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/pause?api-version=2025-04-01-preview

주어진 미세 조정 ID에 의해 지정된 미세 조정 작업 처리를 일시정지하는 것.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
fine-tuning-job-id path Yes string 미세 조정 작업의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 미세 조정이 성공적으로 일시정지되었습니다.

Content-Type Type Description
application/json FineTuningJob

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/pause?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "hyperparameters": {
      "n_epochs": -1
    },
    "integrations": [
      {
        "type": "wandb",
        "wandb": {
          "project": "custom-wandb-project",
          "tags": [
            "project:tag",
            "lineage"
          ]
        }
      }
    ],
    "method": {
      "type": "supervised"
    },
    "status": "paused",
    "model": "curie",
    "training_file": "file-181a1cbdcdcf4677ada87f63a0928099",
    "finished_at": 1646127311,
    "id": "ft-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "fine_tuning.job"
  }
}

미세 조정 - 이력서

POST {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/resume?api-version=2025-04-01-preview

주어진 미세 조정 ID에 의해 지정된 미세 조정 작업 처리를 재개합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
fine-tuning-job-id path Yes string 미세 조정 작업의 식별자입니다.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

설명: 미세 조정이 성공적으로 재개되었습니다.

Content-Type Type Description
application/json FineTuningJob

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/fine_tuning/jobs/{fine-tuning-job-id}/resume?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "hyperparameters": {
      "n_epochs": -1
    },
    "integrations": [
      {
        "type": "wandb",
        "wandb": {
          "project": "custom-wandb-project",
          "tags": [
            "project:tag",
            "lineage"
          ]
        }
      }
    ],
    "method": {
      "type": "supervised"
    },
    "status": "resuming",
    "model": "curie",
    "training_file": "file-181a1cbdcdcf4677ada87f63a0928099",
    "finished_at": 1646127311,
    "id": "ft-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "fine_tuning.job"
  }
}

섭취 일자리 - 목록

GET {endpoint}/openai/ingestion/jobs?api-version=2025-04-01-preview

섭취 작업 목록이 있습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.
mgmt-user-token header No string 작업 공간에 접근하는 데 사용되는 토큰(사용자 컴퓨트 작업에만 필요함).

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json IngestionJobList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/ingestion/jobs?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "body": {
    "value": [
      {
        "jobId": "usercompute-ingestion-job",
        "kind": "UserCompute",
        "workspaceId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev"
      },
      {
        "jobId": "syscompute-ingestion-job",
        "kind": "SystemCompute"
      }
    ]
  }
}

인제션 작업 - 생성

PUT {endpoint}/openai/ingestion/jobs/{job-id}?api-version=2025-04-01-preview

지정된 작업 ID로 인제스티션 작업을 생성합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.
job-id path Yes string 생성될 직업의 ID.
mgmt-user-token header No string 작업 공간에 접근하는 데 사용되는 토큰(사용자 컴퓨트 작업에만 필요함).
aml-user-token header No string 작업 공간 내 작업 내 자원에 접근하는 토큰(사용자 컴퓨트 작업에만 필요함).

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: application/json

Name Type Description Required Default
jobId string No
kind IngestionJobType 직업 유형. Yes

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json IngestionJob

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

PUT {endpoint}/openai/ingestion/jobs/{job-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {
    "operation-location": "https://aoairesource.openai.azure.com/openai/ingestion/jobs/ingestion-job/runs/72a2792ef7d24ba7b82c7fe4a37e379f?api-version=2025-04-01-preview"
  },
  "body": {
    "kind": "SystemCompute",
    "jobId": "ingestion-job",
    "searchServiceConnection": {
      "kind": "EndpointWithManagedIdentity",
      "endpoint": "https://aykame-dev-search.search.windows.net"
    },
    "datasource": {
      "kind": "Storage",
      "connection": {
        "kind": "EndpointWithManagedIdentity",
        "endpoint": "https://mystorage.blob.core.windows.net/",
        "resourceId": "/subscriptions/1234567-abcd-1234-5678-1234abcd/resourceGroups/my-resource/providers/Microsoft.Storage/storageAccounts/mystorage"
      },
      "containerName": "container",
      "chunking": {
        "maxChunkSizeInTokens": 2048
      },
      "embeddings": [
        {
          "connection": {
            "kind": "RelativeConnection"
          },
          "deploymentName": "Ada"
        }
      ]
    },
    "dataRefreshIntervalInHours": 24,
    "completionAction": "keepAllAssets"
  }
}
PUT {endpoint}/openai/ingestion/jobs/{job-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {
    "operation-location": "https://aoairesource.openai.azure.com/openai/ingestion/jobs/ingestion-job/runs/72a2792ef7d24ba7b82c7fe4a37e379f?api-version=2025-04-01-preview"
  },
  "body": {
    "kind": "UserCompute",
    "jobId": "ingestion-job",
    "workspaceId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev",
    "compute": {
      "kind": "ServerlessCompute"
    },
    "target": {
      "kind": "AzureAISearch",
      "connectionId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev/connections/search-connection"
    },
    "datasource": {
      "kind": "Dataset",
      "datasetId": "azureml://locations/centraluseuap/workspaces/83317fe6-efa6-4e4a-b020-d0edd11ec382/data/PlainText/versions/1",
      "datasetType": "uri_folder"
    }
  }
}

섭취 작업 - 받기

GET {endpoint}/openai/ingestion/jobs/{job-id}?api-version=2025-04-01-preview

지정된 직무 ID의 세부 정보를 얻습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.
job-id path Yes string 일의 본질.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json IngestionJob

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/ingestion/jobs/{job-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "body": {
    "kind": "UserCompute",
    "jobId": "ingestion-job",
    "workspaceId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev",
    "compute": {
      "kind": "ServerlessCompute"
    },
    "target": {
      "kind": "AzureAISearch",
      "connectionId": "/subscriptions/f375b912-331c-4fc5-8e9f-2d7205e3e036/resourceGroups/adrama-copilot-demo/providers/Microsoft.MachineLearningServices/workspaces/adrama-rag-dev/connections/search-connection"
    },
    "datasource": {
      "kind": "Dataset",
      "datasetId": "azureml://locations/centraluseuap/workspaces/83317fe6-efa6-4e4a-b020-d0edd11ec382/data/PlainText/versions/1",
      "datasetType": "uri_folder"
    }
  }
}

인지션jobruns - 목록

GET {endpoint}/openai/ingestion/jobs/{job-id}/runs?api-version=2025-04-01-preview

지정된 작업 ID의 실행 횟수를 나열합니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.
job-id path Yes string 일의 본질.
mgmt-user-token header No string 작업 공간에 접근하는 데 사용되는 토큰(사용자 컴퓨트 작업에만 필요함).

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json IngestionJobRunList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/ingestion/jobs/{job-id}/runs?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "body": {
    "value": [
      {
        "jobId": "ingestion-job",
        "runId": "f375b912-331c-4fc5-8e9f-2d7205e3e036",
        "status": "succeeded"
      },
      {
        "jobId": "ingestion-job",
        "runId": "07f2d192-add7-4202-a2e3-858c2577f4fd",
        "status": "failed"
      },
      {
        "jobId": "ingestion-job",
        "runId": "5ef7a436-1147-4cbb-82e0-3d502bcc6a7b",
        "status": "running"
      }
    ]
  }
}

인지션잡런스 - Get

GET {endpoint}/openai/ingestion/jobs/{job-id}/runs/{run-id}?api-version=2025-04-01-preview

지정된 실행 ID의 세부 정보를 지정된 작업 ID의 일부로 받습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.
job-id path Yes string 일의 본질.
run-id path Yes string 이 시리즈의 본인.
mgmt-user-token header No string 작업 공간에 접근하는 데 사용되는 토큰(사용자 컴퓨트 작업에만 필요함).

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json IngestionJobRun

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/ingestion/jobs/{job-id}/runs/{run-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "body": {
    "jobId": "ingestion-job",
    "runId": "5ef7a436-1147-4cbb-82e0-3d502bcc6a7b",
    "status": "running",
    "progress": {
      "stageProgress": [
        {
          "name": "Preprocessing",
          "totalItems": 14,
          "processedItems": 0,
          "state": "notRunning"
        },
        {
          "name": "Indexing",
          "state": "notRunning"
        }
      ]
    }
  }
}

모델 - 목록

GET {endpoint}/openai/models?api-version=2025-04-01-preview

Azure OpenAI 리소스에서 접근 가능한 모든 모델 목록을 제공합니다.

여기에는 기본 모델뿐만 아니라 Azure OpenAI 리소스가 소유한 모든 성공적으로 완성된 미세 조정 모델이 포함됩니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json ModelList

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/models?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "data": [
      {
        "status": "succeeded",
        "capabilities": {
          "fine_tune": true,
          "inference": true,
          "completion": true,
          "chat_completion": false,
          "embeddings": false
        },
        "lifecycle_status": `generally-available`,
        "deprecation": {
          "fine_tune": 1677662127,
          "inference": 1709284527
        },
        "id": "curie",
        "created_at": 1646126127,
        "object": "model"
      },
      {
        "status": "succeeded",
        "model": "curie",
        "fine_tune": "ft-72a2792ef7d24ba7b82c7fe4a37e379f",
        "capabilities": {
          "fine_tune": false,
          "inference": true,
          "completion": true,
          "chat_completion": false,
          "embeddings": false
        },
        "lifecycle_status": `generally-available`,
        "deprecation": {
          "inference": 1709284527
        },
        "id": "curie.ft-72a2792ef7d24ba7b82c7fe4a37e379f",
        "created_at": 1646126127,
        "object": "model"
      }
    ],
    "object": "list"
  }
}

모델 - Get

GET {endpoint}/openai/models/{model-id}?api-version=2025-04-01-preview

주어진 modelId가 지정한 모델에 대한 세부 정보를 얻습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
model-id path Yes string 모델의 식별자.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json Model

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

GET {endpoint}/openai/models/{model-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "status": "succeeded",
    "capabilities": {
      "fine_tune": true,
      "inference": true,
      "completion": true,
      "chat_completion": false,
      "embeddings": false
    },
    "lifecycle_status": `generally-available`,
    "deprecation": {
      "fine_tune": 1677662127,
      "inference": 1709284527
    },
    "id": "curie",
    "created_at": 1646126127,
    "object": "model"
  }
}
GET {endpoint}/openai/models/{model-id}?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "status": "succeeded",
    "model": "curie",
    "fine_tune": "ft-72a2792ef7d24ba7b82c7fe4a37e379f",
    "capabilities": {
      "fine_tune": false,
      "inference": true,
      "completion": true,
      "chat_completion": false,
      "embeddings": false
    },
    "lifecycle_status": `generally-available`,
    "deprecation": {
      "inference": 1709284527
    },
    "id": "curie.ft-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "model"
  }
}

업로드 파일 - 시작

POST {endpoint}/openai/uploads?api-version=2025-04-01-preview

중간 업로드 객체가 생성되어 Parts를 추가할 수 있습니다. 현재 업로드 용량은 최대 9GB이며, 생성 후 2시간 이내에 만료됩니다.

업로드가 완료되면 업로드된 모든 부분을 포함하는 파일 객체가 생성됩니다. 이 파일 객체는 다른 파일처럼 우리 플랫폼 전반에서 사용할 수 있습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: application/json

Name Type Description Required Default
bytes integer Yes
filename string Yes
mime_type string Yes
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes

Request Body

Content-Type: text/json

Name Type Description Required Default
bytes integer Yes
filename string Yes
mime_type string Yes
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes

Request Body

Content-Type: application/*+json

Name Type Description Required Default
bytes integer Yes
filename string Yes
mime_type string Yes
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes

Responses

상태 코드: 200

설명: 업로드가 성공적으로 생성되었습니다.

Content-Type Type Description
application/json UploadResource

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/uploads?api-version=2025-04-01-preview

{
 "purpose": "fine-tune",
 "filename": "training_data_v21.jsonl",
 "bytes": 2097152,
 "mime_type": "application/json"
}

응답: 상태 코드: 200

{
  "headers": {
    "location": "https://aoairesource.openai.azure.com/openai/uploads/runs/fine-tune-72a2792ef7d24ba7b82c7fe4a37e379f"
  },
  "body": {
    "bytes": 2097152,
    "filename": "training_data_v21.jsonl",
    "purpose": "fine-tune",
    "status": "pending",
    "expires_at": 1646133327,
    "file": {
      "status": "pending",
      "bytes": 140,
      "purpose": "fine-tune",
      "filename": "puppy.jsonl",
      "id": "file-181a1cbdcdcf4677ada87f63a0928099",
      "created_at": 1646126127,
      "object": "file"
    },
    "id": "fine-tune-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "upload"
  }
}

파일 업로드 - 취소

POST {endpoint}/openai/uploads/{upload-id}/cancel?api-version=2025-04-01-preview

업로드를 취소하고, 업로드된 모든 부품이 비동기적으로 삭제됩니다.

업로드가 취소된 후에는 어떤 부품도 추가할 수 없습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
upload-id path Yes string 업로드 식별자.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json UploadResource

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/uploads/{upload-id}/cancel?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "bytes": 2097152,
    "filename": "training_data_v21.jsonl",
    "purpose": "fine-tune",
    "status": "cancelled",
    "expires_at": 1646133327,
    "file": {
      "status": "pending",
      "bytes": 140,
      "purpose": "fine-tune",
      "filename": "puppy.jsonl",
      "id": "file-181a1cbdcdcf4677ada87f63a0928099",
      "created_at": 1646126127,
      "object": "file"
    },
    "id": "fine-tune-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "upload"
  }
}

업로드 파일 - 완료

POST {endpoint}/openai/uploads/{upload-id}/complete?api-version=2025-04-01-preview

이렇게 하면 업로드가 완료되고, 반환된 업로드 객체는 플랫폼 전반에 걸쳐 사용할 준비가 된 중첩된 파일 객체를 포함합니다.

부품 ID의 순서 목록을 제공함으로써 부품 순서를 정의할 수 있습니다.

업로드된 총 바이트 수는 업로드 객체를 생성할 때 처음 지정한 크기와 일치해야 합니다.

이 작업 이후에는 업로드가 완료된 후 추가 부품을 추가할 수 없습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
upload-id path Yes string 업로드 식별자.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: application/json

Name Type Description Required Default
md5 string No
part_ids array Yes

Request Body

Content-Type: text/json

Name Type Description Required Default
md5 string No
part_ids array Yes

Request Body

Content-Type: application/*+json

Name Type Description Required Default
md5 string No
part_ids array Yes

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json UploadResource

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/uploads/{upload-id}/complete?api-version=2025-04-01-preview

{
 "part_ids": [
  "LnmictL3p0u4LH/ko343nw==",
  "LmmictL3p0u4LH/ko343nw=="
 ]
}

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "bytes": 2097152,
    "filename": "training_data_v21.jsonl",
    "purpose": "fine-tune",
    "status": "completed",
    "expires_at": 1646133327,
    "file": {
      "status": "processed",
      "bytes": 140,
      "purpose": "fine-tune",
      "filename": "puppy.jsonl",
      "id": "file-181a1cbdcdcf4677ada87f63a0928099",
      "created_at": 1646126127,
      "object": "file"
    },
    "id": "fine-tune-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "upload"
  }
}

업로드 파일 - 부품

POST {endpoint}/openai/uploads/{upload-id}/parts?api-version=2025-04-01-preview

업로드 객체에 부품을 추가하며, 각 부품은 업로드하는 파일의 한 구간을 나타냅니다.

각 파트는 파일 목적에 따라 표준 파일 업로드 크기 제한까지 설정할 수 있습니다. 업로드 용량 9GB에 도달할 때까지 부품을 계속 추가할 수 있습니다.

URI Parameters

Name In Required Type Description
endpoint path Yes string
url
OpenAI 엔드포인트Azure 지원(프로토콜 및 호스트명, 예: https://aoairesource.openai.azure.com. "aoairesource"를 Azure OpenAI 리소스 이름으로 대체하세요. https://{your-resource-name}.openai.azure.com
upload-id path Yes string 업로드 식별자.
api-version query Yes string 요청된 API 버전입니다.

Request Header

토큰 기반 인증이나 API 키를 사용하세요. 토큰 기반 인증을 사용하는 것이 권장되며 더 안전합니다.

Name Required Type Description
Authorization True string Example:Authorization: Bearer {Azure_OpenAI_Auth_Token}

Azure CLI 사용해 인증 토큰을 생성하려면 az account get-access-token --resource https://cognitiveservices.azure.com

Type: oauth2
인증 URL: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
범위: https://ai.azure.com/.default
api-key True string Provide Azure OpenAI API key here

Request Body

Content-Type: multipart/form-data

Name Type Description Required Default
data string Yes

Responses

상태 코드: 200

Description: Success

Content-Type Type Description
application/json UploadPartResource

상태 코드: 기본 상태

설명: 오류가 발생했습니다.

Content-Type Type Description
application/json ErrorResponse

Examples

POST {endpoint}/openai/uploads/{upload-id}/parts?api-version=2025-04-01-preview

응답: 상태 코드: 200

{
  "headers": {},
  "body": {
    "upload_id": "fine-tune-72a2792ef7d24ba7b82c7fe4a37e379f",
    "azure_block_id": "LnmictL3p0u4LH/ko343nw==",
    "id": "uplprt-72a2792ef7d24ba7b82c7fe4a37e379f",
    "created_at": 1646126127,
    "object": "upload.part"
  }
}

Components

AzureAISearchIndex

Azure AI 검색 인덱스.

Name Type Description Required Default
connectionId string 연결 ID는 Azure AI 검색 Index를 가리키고 있습니다. No
kind TargetType 타깃 유형. Yes

BaseConnection

자원과의 연결고리.

BaseConnection을 위한 판별기

이 구성 요소는 다음과 같은 속성을 kind 사용하여 서로 다른 유형을 구별합니다:

Type Value Schema
Name Type Description Required Default
kind ConnectionType 연결 유형. Yes

Batch

배치의 값을 정의합니다.

Name Type Description Required Default
cancelled_at integer 이 배치가 취소된 타임스탬프(유닉스 에포흐 기준)입니다. No
cancelling_at integer 이 배치가 취소되기 시작한 타임스탬프(유닉스 에포크)입니다. No
completed_at integer 이 배치가 완료된 타임스탬프(유닉스 에포크 기준)입니다. No
completion_window string 배치가 처리되어야 하는 시간 프레임. No
created_at integer 이 배치가 생성된 타임스탬프(유닉스 에포흐 기준)입니다. No
endpoint string 배치가 사용하는 API 엔드포인트입니다. No
error_blob string 오류가 있는 요청의 출력을 담은 블롭 URL. No
error_file_id string 오류가 있는 요청의 출력을 포함하는 파일의 ID입니다. No
errors BatchErrors 실패한 배치의 경우, 여기에는 실패 원인에 대한 추가 정보가 포함됩니다. No
expired_at integer 이 배치가 만료된 타임스탬프(유닉스 에포크 기준)입니다. No
expires_at integer 이 배치가 만료되는 타임스탬프(유닉스 에포크)입니다. No
failed_at integer 이 배치가 실패한 타임스탬프(유닉스 에포크)입니다. No
finalizing_at integer 이 배치가 최종 확정을 시작한 타임스탬프(유닉스 에포크)입니다. No
id string 이 물건의 정체. Yes
in_progress_at integer 이 배치가 진행되기 시작한 타임스탬프(유닉스 에포크)입니다. No
input_blob string 배치 입력 파일이 포함된 블롭 URL. No
input_file_id string 배치의 입력 파일의 ID입니다. Yes
metadata object 배치에 부착할 수 있는 키-값 쌍 집합입니다. 이는 배치에 대한 추가 정보를 구조화된 형식으로 저장하는 데 유용할 수 있습니다. No
object TypeDiscriminator 객체의 유형을 정의합니다. No
output_blob string 성공적으로 실행된 요청의 출력을 담은 블롭 URL. No
output_file_id string 성공적으로 실행된 요청의 출력을 포함하는 파일의 ID. No
request_counts BatchRequestCounts 요청은 배치 내 다양한 상태에 따라 계산됩니다. No
status BatchStatus 배치의 상태. No

BatchCreateRequest

배치 생성 요청을 정의합니다.

Name Type Description Required Default
completion_window string 배치가 처리되어야 하는 시간 프레임. Yes
endpoint string 배치가 사용하는 API 엔드포인트입니다. Yes
input_blob string 배치 입력으로 사용할 Azure Storage 블롭의 URL. No
input_file_id string 배치의 입력 파일의 ID입니다. No
metadata object 배치에 부착할 수 있는 키-값 쌍 집합입니다. 이는 배치에 대한 추가 정보를 구조화된 형식으로 저장하는 데 유용할 수 있습니다. No
output_expires_after FileExpiresAfter 파일의 만료일을 정의합니다. No
output_folder BatchOutputReference 출력을 저장하는 Azure Storage 폴더. No

BatchErrorData

배치 실패에 대한 오류 정보입니다.

Name Type Description Required Default
code string 오류 유형을 식별하는 오류 코드입니다. No
line string 오류가 발생한 입력 파일의 줄 번호(해당되는 경우)를 표시합니다(null일 수도 있음). No
message string 오류가 발생한 내용에 대한 더 자세한 내용을 담은 사람이 읽을 수 있는 메시지입니다. No
param string 오류를 일으킨 매개변수의 이름(해당된다면) (null일 수도 있음)을 알려줍니다. No

BatchErrors

실패한 배치의 경우, 여기에는 실패 원인에 대한 추가 정보가 포함됩니다.

Name Type Description Required Default
data BatchErrorData 배치 실패에 대한 오류 정보입니다. No
object string 오류 객체의 유형입니다. 항상 '리스트'입니다. No

BatchRequestCounts

요청은 배치 내 다양한 상태에 따라 계산됩니다.

Name Type Description Required Default
completed integer 배치 내에서 성공적으로 완료된 요청 수. No
failed integer 배치 내에서 실패한 요청의 수. No
total integer 배치 내 전체 요청 수. No

BatchOutputReference

출력을 저장하는 Azure Storage 폴더.

Name Type Description Required Default
delimiter string Optional. 기본적으로 폴더 경로에 사용되는 구분자입니다 /. No
url string 배치 출력물이 저장될 Azure Storage 폴더의 URL입니다. No

BatchStatus

배치의 상태.

Property Value
Description 배치의 상태.
Type string
Values validating
failed
in_progress
finalizing
completed
expired
cancelling
cancelled

BatchesList

배치 목록입니다.

Name Type Description Required Default
data array 물건 목록. No
first_id string 배치 목록에서 첫 번째 배치의 ID가 반환되었습니다. No
has_more boolean 리스트에 반환된 것보다 더 많은 요소를 포함하고 있는지를 나타내는 값입니다. No
last_id string 배치 목록에서 마지막 배치의 ID가 반환되었습니다. No
object TypeDiscriminator 객체의 유형을 정의합니다. No

Capabilities

기본 모델이나 미세 조정 모델의 기능.

Name Type Description Required Default
chat_completion boolean 이 값은 모델이 채팅 완료를 지원하는지 여부를 나타냅니다. Yes
completion boolean 모델이 완성을 지원하는지 여부를 나타내는 값입니다. Yes
embeddings boolean 모델이 임베딩을 지원하는지 여부를 나타내는 값입니다. Yes
fine_tune boolean 모델이 미세 조정에 사용될 수 있는지를 나타내는 값입니다. Yes
inference boolean 모델이 배포 가능한지 여부를 나타내는 값입니다. Yes

Choice

Name Type Description Required Default
finish_reason string 모델이 토큰 생성을 중단한 이유입니다. No
index integer 선택 목록 내 선택의 인덱스입니다. No
logprobs object 선택에 대한 확률 정보를 기록하세요. No
message ChoiceMessage No

ChoiceMessage

Name Type Description Required Default
content string 메시지 내용. No
role string 메시지를 생성하는 주체의 역할. No
tool_calls array 관련 툴 호출 목록입니다. No

ChunkingSettings

Chunking settings

Name Type Description Required Default
maxChunkSizeInTokens integer No

CompletionAction

완성 동작.

Property Value
Description 완성 동작.
Type string
Values cleanUpTempAssets
keepAllAssets

ComputeType

컴퓨트 타입입니다.

Property Value
Description 컴퓨트 타입입니다.
Type string
Values ServerlessCompute
CustomCompute

ConnectionStringConnection

연결 끈 연결.

Name Type Description Required Default
connectionString string Connection string No
kind ConnectionType 연결 유형. Yes

ConnectionType

연결 유형.

Property Value
Description 연결 유형.
Type string
Values EndpointWithKey
ConnectionString
EndpointWithManagedIdentity
WorkspaceConnection
RelativeConnection

CosmosDBIndex

CosmosDB Index.

Name Type Description Required Default
collectionName string 코스모스 DB 컬렉션의 이름입니다. No
connectionId string 연결 ID가 코스모스 DB를 가리키고 있습니다. No
databaseName string 코스모스 DB 데이터베이스의 이름입니다. No
kind TargetType 타깃 유형. Yes

CrawlingSettings

Crawling settings

Name Type Description Required Default
maxCrawlDepth integer No
maxCrawlTimeInMins integer No
maxDownloadTimeInMins integer No
maxFileSize integer No
maxFiles integer No
maxRedirects integer No

CreateEvalCompletionsRunDataSource

모델 샘플링 구성을 설명하는 CompletionsRunDataSource 객체입니다.

Name Type Description Required Default
input_messages object No
└─ item_reference string '항목' 네임스페이스의 변수, 예를 들어 'item.name'에 대한 참조입니다. No
└─ template array 프롬프트나 맥락을 구성하는 채팅 메시지 목록입니다. '항목' 네임스페이스에 대한 변수 참조를 포함할 수 있습니다. 예를 들어 {{item.name}}. No
└─ type enum 입력 메시지의 종류. 또는 templateitem_reference.
가능한 값들: template, item_reference
No
model string 완성 생성에 사용할 모델명(예: 'o3-mini')입니다. No
sampling_params object No
└─ max_completion_tokens integer 생성된 출력에 포함된 최대 토큰 수. No
└─ seed integer 샘플링 중 무작위성을 초기화하는 시드 값입니다. No 42
└─ temperature number 온도가 높을수록 출력의 무작위성이 증가합니다. No 1
└─ top_p number 원자핵 샘플링을 위한 온도 대안; 1.0은 모든 토큰을 포함합니다. No 1
source object Yes
└─ type enum 출처의 종류. , , EvalJsonlFileIdSourceEvalStoredCompletionsSource또는 중 하나EvalJsonlFileContentSource일 수 있습니다.
가능한 값들: EvalJsonlFileContentSource, EvalJsonlFileIdSource, EvalStoredCompletionsSource
No
type enum 런 데이터 소스의 유형. 항상 그렇습니다.completions
가능한 값: completions
Yes

CreateEvalCustomDataSourceConfig

평가 실행에 사용되는 데이터 소스의 스키마를 정의하는 CustomDataSourceConfig 객체입니다. 이 스키마는 다음 데이터 형태를 정의하는 데 사용됩니다: - 테스트 기준을 정의하는 데 사용되고 - 런을 생성할 때 필요한 데이터가 무엇인지

Name Type Description Required Default
include_sample_schema boolean 평가가 샘플 네임스페이스를 채우는 것을 기대해야 하는지(즉, 데이터 소스에서 응답을 생성하는 것) No False
item_schema object 데이터 소스의 각 행에 대한 json 스키마입니다. Yes
type enum 데이터 소스의 종류. 항상 그렇습니다.custom
가능한 값: custom
Yes

CreateEvalItem

프롬프트나 맥락을 구성하는 채팅 메시지입니다. "item" 네임스페이스에 대한 변수 참조, 예를 들어 {{item.name}}를 포함할 수 있습니다.

Name Type Description Required Default
content ['string', 'InputTextContent', 'object'] 모델에 대한 텍스트 입력은 템플릿 문자열을 포함할 수 있습니다. Yes
role enum 메시지 입력의 역할. , userassistant, system, , 중 developer하나 중 하나 .
가능한 값들: user, assistant, system, developer
Yes
type enum 메시지 입력의 유형. 항상 그렇습니다.message
가능한 값: message
No

CreateEvalJsonlRunDataSource

JsonlRunDataSource 객체를 통해 평가와 일치하는 JSONL 파일을 지정합니다

Name Type Description Required Default
source object Yes
└─ content array jsonl 파일의 내용입니다. No
└─ id string 파일의 식별자입니다. No
└─ type enum jsonl 소스 유형. 항상 그렇습니다.file_id
가능한 값: file_id
No
type enum 데이터 소스의 종류. 항상 그렇습니다.jsonl
가능한 값: jsonl
Yes

CreateEvalLabelModelGrader

평가 내 각 항목에 라벨을 할당하는 모델을 사용하는 LabelModelGrader 객체입니다.

Name Type Description Required Default
input array 프롬프트나 맥락을 구성하는 채팅 메시지 목록입니다. "item" 네임스페이스에 대한 변수 참조, 예를 들어 {{item.name}}를 포함할 수 있습니다. Yes
labels array 평가 항목별로 분류할 라벨. Yes
model string 평가에 사용할 모델. 구조화된 산출물을 지원해야 합니다. Yes
name string 채점자의 이름입니다. Yes
passing_labels array 합격 결과를 나타내는 라벨들. 라벨의 일부일 거야. Yes
type enum 객체 유형은 항상 label_model입니다.
가능한 값: label_model
Yes

CreateEvalLogsDataSourceConfig

저장된 완료 쿼리의 메타데이터 속성을 지정하는 데이터 소스 설정입니다. 보통 이런 메타데이터 usecase=chatbot 는 또는 prompt-version=v2,

Name Type Description Required Default
metadata object 로그 데이터 소스에 대한 메타데이터 필터. No
type enum 데이터 소스의 종류. 항상 그렇습니다.logs
가능한 값: logs
Yes

CreateEvalRequest

Name Type Description Required Default
data_source_config object 평가 실행에 사용되는 데이터 소스의 스키마를 정의하는 CustomDataSourceConfig 객체입니다. 이 스키마는 다음 데이터 형태를 정의하는 데 사용됩니다: - 테스트 기준을 정의하는 데 사용되고 - 런을 생성할 때 필요한 데이터가 무엇인지 Yes
└─ include_sample_schema boolean 평가가 샘플 네임스페이스를 채우는 것을 기대해야 하는지(즉, 데이터 소스에서 응답을 생성하는 것) No False
└─ item_schema object 데이터 소스의 각 행에 대한 json 스키마입니다. No
└─ metadata object 로그 데이터 소스에 대한 메타데이터 필터. No
└─ type enum 데이터 소스의 종류. 항상 그렇습니다.logs
가능한 값: logs
No
metadata Metadata 객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다. No
name string 평가 이름. No
testing_criteria array 이 그룹의 모든 평가 구간의 채점 목록입니다. Yes

CreateEvalRunRequest

Name Type Description Required Default
data_source object JsonlRunDataSource 객체를 통해 평가와 일치하는 JSONL 파일을 지정합니다 Yes
└─ input_messages object No
└─ item_reference string '항목' 네임스페이스의 변수, 예를 들어 'item.name'에 대한 참조입니다. No
└─ template array 프롬프트나 맥락을 구성하는 채팅 메시지 목록입니다. '항목' 네임스페이스에 대한 변수 참조를 포함할 수 있습니다. 예를 들어 {{item.name}}. No
└─ type enum 입력 메시지의 종류. 또는 templateitem_reference.
가능한 값들: template, item_reference
No
└─ model string 완성 생성에 사용할 모델명(예: 'o3-mini')입니다. No
└─ sampling_params object No
└─ max_completion_tokens integer 생성된 출력에 포함된 최대 토큰 수. No
└─ seed integer 샘플링 중 무작위성을 초기화하는 시드 값입니다. No 42
└─ temperature number 온도가 높을수록 출력의 무작위성이 증가합니다. No 1
└─ top_p number 원자핵 샘플링을 위한 온도 대안; 1.0은 모든 토큰을 포함합니다. No 1
└─ source object No
└─ type enum 출처의 종류. , , EvalJsonlFileIdSourceEvalStoredCompletionsSource또는 중 하나EvalJsonlFileContentSource일 수 있습니다.
가능한 값들: EvalJsonlFileContentSource, EvalJsonlFileIdSource, EvalStoredCompletionsSource
No
└─ type enum 런 데이터 소스의 유형. 항상 그렇습니다.completions
가능한 값: completions
No
metadata Metadata 객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다. No
name string 그 런의 이름. No

CreateEvalStoredCompletionsRunDataSource

필터 집합을 설명하는 StoredCompletionsRunDataSource 구성

Name Type Description Required Default
created_after integer 이 시기 이후에 생성된 항목을 필터링할 수 있는 선택적 유닉스 타임스탬프가 있습니다. No
created_before integer 이 시기 이전에 생성된 항목을 필터링하는 선택적 Unix 타임스탬프가 있습니다. No
limit integer 반품할 수 있는 최대 아이템 수는 선택 사항입니다. No
metadata Metadata 객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다. No
model string 필터링할 수 있는 선택적 모델(예: 'gpt-4o')이 있습니다. No
type enum 데이터 소스의 종류. 항상 그렇습니다.stored_completions
가능한 값: stored_completions
Yes

CustomCompute

Custom compute.

Name Type Description Required Default
computeId string 커스텀 컴퓨트의 ID No
kind ComputeType 컴퓨트 타입입니다. Yes

DeploymentConnection

상대적 배치 연결.

Name Type Description Required Default
kind ConnectionType 연결 유형. Yes

Deprecation

모델의 다양한 사용 사례에 대한 폐기일을 정의합니다.

보통 기본 모델은 제작 후 1년간 미세 조정을 지원합니다. 추론은 일반적으로 기본 모델이나 미세 조정 모델 생성 후 2년 후에 지원됩니다. 정확한 날짜는 부동산에 명시되어 있습니다.

Name Type Description Required Default
fine_tune integer 이 모델의 미세 조정 지원이 종료된 날짜입니다. 미세 조정 모델용입니다 null . No
inference integer 이 모델의 추론 지원 종료일. Yes

DpoHyperparamsRequest

DPO 미세 조정 방법의 하이퍼파라미터

Name Type Description Required Default
batch_size integer 학습에 사용할 일괄 처리 크기입니다. 일괄 처리 크기는 단일 앞으로 및 뒤로 패스를 학습하는 데 사용되는 학습 예제의 수입니다. 일반적으로 우리는 더 큰 배치 크기가 더 큰 데이터셋에 더 잘 작동하는 경향이 있음을 발견했습니다. 이 속성의 기본값과 최대 값은 기본 모델에 따라 다릅니다. No
beta number DPO 베타 값. No
l2_multiplier number L2 정규화 승수기 No
learning_rate_multiplier number 학습에 사용할 학습 속도 승수입니다. 미세 조정 학습률은 사전 학습에 사용된 원래 학습률에 이 값을 곱한 값입니다. 학습 속도가 클수록 일괄 처리 크기가 클수록 성능이 향상되는 경향이 있습니다. 0.02에서 0.2 범위의 값을 실험해 보시고 어떤 값이 가장 좋은 결과를 내는지 확인해 보시길 권장합니다. No
n_epochs integer 모델을 학습시킬 Epoch의 수입니다. Epoch는 학습 데이터 세트를 통해 하나의 전체 주기를 나타냅니다. No

DpoMethod

DPO 메서드 미세 조정 요청

Name Type Description Required Default
dpo DpoMethodConfigRequest 미세 조정 방법의 dpo 구성. DPO 특화 하이퍼파라미터 포함 Yes
type FineTuningMethodType Yes

DpoMethodConfigRequest

미세 조정 방법의 dpo 구성. DPO 특화 하이퍼파라미터 포함

Name Type Description Required Default
hyperparameters DpoHyperparamsRequest DPO 미세 조정 방법의 하이퍼파라미터 No

DpoMethodRequest

DPO 메서드 미세 조정 요청

Name Type Description Required Default
dpo DpoMethodConfigRequest 미세 조정 방법의 dpo 구성. DPO 특화 하이퍼파라미터 포함 Yes
type FineTuningMethodType Yes

EndpointKeyConnection

엔드포인트 키 연결.

Name Type Description Required Default
endpoint string Endpoint No
key string Key No
kind ConnectionType 연결 유형. Yes

EndpointMIConnection

엔드포인트 관리 신원 연결.

Name Type Description Required Default
endpoint string Endpoint No
kind ConnectionType 연결 유형. Yes
resourceId string Resource Id No

Error

오류 내용은 Microsoft REST 가이드라인에 정의된 것입니다.

Name Type Description Required Default
code ErrorCode Microsoft REST 가이드라인에 정의된 오류 코드. Yes
details array 오류 정보가 있으면 알려주세요. No
innererror InnerError 내부 오류는 Microsoft REST 가이드라인에 정의된 것입니다. No
message string 이 오류의 메시지. Yes
target string 오류가 발생한 위치가 가능하다면. No

ErrorCode

Microsoft REST 가이드라인에 정의된 오류 코드.

Property Value
Description Microsoft REST 가이드라인에 정의된 오류 코드.
Type string
Values conflict
invalidPayload
forbidden
notFound
unexpectedEntityState
itemDoesAlreadyExist
serviceUnavailable
internalFailure
quotaExceeded
jsonlValidationFailed
fileImportFailed
tooManyRequests
unauthorized
contentFilter

ErrorResponse

Microsoft REST 가이드라인에 정의된 오류 응답.

Name Type Description Required Default
error Error 오류 내용은 Microsoft REST 가이드라인에 정의된 것입니다. Yes

Eval

데이터 소스, 설정, 테스트 기준이 포함된 평가 객체입니다. 평가는 LLM 통합을 위해 수행해야 할 작업을 나타냅니다. Like:

  • 내 챗봇의 품질을 개선해
  • 내 챗봇이 고객 지원을 얼마나 잘 처리하는지 확인해 보세요
  • o3-mini가 GPT-4o보다 제 사용 상황에 더 적합한지 확인해 보세요
Name Type Description Required Default
created_at integer 평가가 생성된 시점의 Unix 타임스탬프(초 단위)입니다. Yes
data_source_config object 네임스페이스와 선택적으로 sample 네임스페이스의 item 스키마를 지정하는 CustomDataSourceConfig입니다. 응답 스키마는 다음과 같은 데이터 형태를 정의합니다: - 테스트 기준을 정의하는 데 사용되고 - 런을 생성할 때 필요한 데이터가 무엇인지가 Yes
└─ metadata Metadata 객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다. No
└─ schema object 실행 데이터 소스 항목에 대한 json 스키마입니다. 여기에서 JSON 스키마 구축 방법을 배워보세요. No
└─ type enum 데이터 소스의 종류. 항상 그렇습니다.stored_completions
가능한 값: stored_completions
No
id string 평가의 고유 식별자. Yes
metadata Metadata 객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다. Yes
name string 평가 이름. Yes
object enum 객체 유형.
가능한 값: eval
Yes
testing_criteria array 검사 기준 목록입니다. Yes

EvalApiError

Eval API의 오류 응답을 나타내는 객체입니다.

Name Type Description Required Default
code string 오류 코드. Yes
message string 오류 메시지. Yes

EvalCustomDataSourceConfig

네임스페이스와 선택적으로 sample 네임스페이스의 item 스키마를 지정하는 CustomDataSourceConfig입니다. 응답 스키마는 다음과 같은 데이터 형태를 정의합니다: - 테스트 기준을 정의하는 데 사용되고 - 런을 생성할 때 필요한 데이터가 무엇인지가

Name Type Description Required Default
schema object 실행 데이터 소스 항목에 대한 json 스키마입니다. 여기에서 JSON 스키마 구축 방법을 배워보세요. Yes
type enum 데이터 소스의 종류. 항상 그렇습니다.custom
가능한 값: custom
Yes

EvalFileRunDataSource

파일 ID가 있는 FileRunDataSource 구성

Name Type Description Required Default
file_id string 파일의 식별자입니다. Yes
type enum 데이터 소스의 종류. 항상 그렇습니다.file
가능한 값: file
Yes

EvalItem

계층 구조 다음 명령을 나타내는 역할이 있는 모델 메시지 입력. 또는 system 역할에 포함된 developer 지시가 해당 역할에 user 주어진 지시보다 우선합니다. 역할을 가진 assistant 메시지는 이전 상호작용에서 모델에 의해 생성된 것으로 추정됩니다.

Name Type Description Required Default
content object 모델에 입력하는 텍스트. Yes
└─ text string 모델에서 출력된 텍스트입니다. No
└─ type enum 출력 텍스트의 유형. 항상 그렇습니다.output_text
가능한 값: output_text
No
role enum 메시지 입력의 역할. , userassistant, system, , 중 developer하나 중 하나 .
가능한 값들: user, assistant, system, developer
Yes
type enum 메시지 입력의 유형. 항상 그렇습니다.message
가능한 값: message
No

EvalJsonlFileContentSource

Name Type Description Required Default
content array jsonl 파일의 내용입니다. Yes
type enum jsonl 소스 유형. 항상 그렇습니다.file_content
가능한 값: file_content
Yes

EvalJsonlFileIdSource

Name Type Description Required Default
id string 파일의 식별자입니다. Yes
type enum jsonl 소스 유형. 항상 그렇습니다.file_id
가능한 값: file_id
Yes

EvalLabelModelGrader

평가 내 각 항목에 라벨을 할당하는 모델을 사용하는 LabelModelGrader 객체입니다.

Name Type Description Required Default
input array Yes
labels array 평가 항목마다 할당할 라벨들. Yes
model string 평가에 사용할 모델. 구조화된 산출물을 지원해야 합니다. Yes
name string 채점자의 이름입니다. Yes
passing_labels array 합격 결과를 나타내는 라벨들. 라벨의 일부일 거야. Yes
type enum 객체 유형은 항상 label_model입니다.
가능한 값: label_model
Yes

EvalList

평가 목록을 나타내는 물체.

Name Type Description Required Default
data array 평가용 물건들이 배열되어 있었다. Yes
first_id string 데이터 배열에서 첫 번째 평가의 식별자입니다. Yes
has_more boolean 더 많은 평가가 있는지 여부를 나타냅니다. Yes
last_id string 데이터 배열에서 마지막 평가의 식별자입니다. Yes
object enum 이 물체의 종류. 항상 "list"로 설정되어 있습니다.
가능한 값: list
Yes

EvalRun

평가 실행을 나타내는 스키마입니다.

Name Type Description Required Default
created_at integer 평가 실행이 생성될 때 유닉스 타임스탬프(초 단위)를 기록했습니다. Yes
data_source object JsonlRunDataSource 객체를 통해 평가와 일치하는 JSONL 파일을 지정합니다 Yes
└─ input_messages object No
└─ item_reference string '항목' 네임스페이스의 변수, 예를 들어 'item.name'에 대한 참조입니다. No
└─ template array 프롬프트나 맥락을 구성하는 채팅 메시지 목록입니다. '항목' 네임스페이스에 대한 변수 참조를 포함할 수 있습니다. 예를 들어 {{item.name}}. No
└─ type enum 입력 메시지의 종류. 또는 templateitem_reference.
가능한 값들: template, item_reference
No
└─ model string 완성 생성에 사용할 모델명(예: 'o3-mini')입니다. No
└─ sampling_params object No
└─ max_completion_tokens integer 생성된 출력에 포함된 최대 토큰 수. No
└─ seed integer 샘플링 중 무작위성을 초기화하는 시드 값입니다. No 42
└─ temperature number 온도가 높을수록 출력의 무작위성이 증가합니다. No 1
└─ top_p number 원자핵 샘플링을 위한 온도 대안; 1.0은 모든 토큰을 포함합니다. No 1
└─ source object No
└─ type enum 출처의 종류. , , EvalJsonlFileIdSourceEvalStoredCompletionsSource또는 중 하나EvalJsonlFileContentSource일 수 있습니다.
가능한 값들: EvalJsonlFileContentSource, EvalJsonlFileIdSource, EvalStoredCompletionsSource
No
└─ type enum 런 데이터 소스의 유형. 항상 그렇습니다.completions
가능한 값: completions
No
error EvalApiError Eval API의 오류 응답을 나타내는 객체입니다. Yes
eval_id string 관련 평가의 식별자. Yes
id string 평가 실행의 고유 식별자. Yes
metadata Metadata 객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다. Yes
model string 해당되는 경우 평가되는 모델입니다. Yes
name string 평가 진행 이름입니다. Yes
object enum 물건의 종류. Always "eval.run".
가능한 값: eval.run
Yes
per_model_usage array 평가 실행 중 각 모델의 사용 통계. Yes
per_testing_criteria_results array 평가 실행 중 적용된 시험 기준에 따른 결과. Yes
report_url string UI 대시보드에서 렌더링된 평가 실행 보고서의 URL을 제공합니다. Yes
result_counts object 평가 결과를 요약한 카운터. Yes
└─ errored integer 오류를 발생시킨 출력 항목 수. No
└─ failed integer 평가에 실패한 출력 항목 수. No
└─ passed integer 평가를 통과한 출력 항목 수. No
└─ total integer 실행 출력 항목의 총 수. No
status string 평가 진행 상태. Yes

EvalRunList

평가를 위한 실행 목록을 나타내는 객체입니다.

Name Type Description Required Default
data array 평가 실행 객체 배열. Yes
first_id string 데이터 배열에서 첫 번째 평가 실행의 식별자입니다. Yes
has_more boolean 더 많은 평가가 있는지 여부를 나타냅니다. Yes
last_id string 데이터 배열에서 마지막 평가 실행의 식별자입니다. Yes
object enum 이 물체의 종류. 항상 "list"로 설정되어 있습니다.
가능한 값: list
Yes

EvalRunOutputItem

평가 실행 출력 항목을 나타내는 스키마입니다.

Name Type Description Required Default
created_at integer 평가 실행이 생성될 때 유닉스 타임스탬프(초 단위)를 기록했습니다. Yes
datasource_item object 입력 데이터 소스 항목의 세부 사항. Yes
datasource_item_id integer 데이터 소스 항목의 식별자입니다. Yes
eval_id string 평가 그룹의 식별자입니다. Yes
id string 평가 실행 출력 항목의 고유 식별자. Yes
object enum 물건의 종류. Always "eval.run.output_item".
가능한 값: eval.run.output_item
Yes
results array 평가 실행 결과 목록입니다. Yes
run_id string 이 출력 항목과 연관된 평가 실행의 식별자입니다. Yes
sample object 평가 실행의 입력과 출력을 포함하는 샘플입니다. Yes
└─ error EvalApiError Eval API의 오류 응답을 나타내는 객체입니다. No
└─ finish_reason string 샘플 생성이 완료된 이유입니다. No
└─ input array 입력 메시지 배열. No
└─ max_completion_tokens integer 완료할 수 있는 최대 토큰 수가 제한되었습니다. No
└─ model string 샘플을 생성하는 데 사용된 모델입니다. No
└─ output array 출력 메시지 배열입니다. No
└─ seed integer 샘플을 생성하는 데 사용되는 씨앗입니다. No
└─ temperature number 샘플링 온도가 사용됩니다. No
└─ top_p number 샘플링에 사용되는 top_p 값입니다. No
└─ usage object 샘플의 토큰 사용 세부사항. No
└─ cached_tokens integer 캐시에서 가져온 토큰 수. No
└─ completion_tokens integer 생성된 완료 토큰의 수. No
└─ prompt_tokens integer 사용된 프롬프트 토큰의 수. No
└─ total_tokens integer 총 사용된 토큰 수. No
status string 평가 진행 상태. Yes

EvalRunOutputItemList

평가 실행을 위한 출력 항목 목록을 나타내는 객체입니다.

Name Type Description Required Default
data array 평가 실행 출력 항목 객체 배열입니다. Yes
first_id string 첫 번째 평가 실행의 식별자는 데이터 배열에서 출력 항목입니다. Yes
has_more boolean 더 많은 평가 실행 출력 항목이 있는지 여부를 나타냅니다. Yes
last_id string 마지막 평가 실행의 식별자가 데이터 배열에서 출력 항목으로 사용됩니다. Yes
object enum 이 물체의 종류. 항상 "list"로 설정되어 있습니다.
가능한 값: list
Yes

EvalScoreModelGrader

모델을 사용해 입력에 점수를 부여하는 ScoreModelGrader 객체입니다.

Name Type Description Required Default
input array 입력된 텍스트. 여기에는 템플릿 문자열이 포함될 수 있습니다. Yes
model string 평가에 사용할 모델. Yes
name string 채점자의 이름입니다. Yes
pass_threshold number 점수의 기준점입니다. No
range array 악보의 범위. 기본값은 [0, 1]입니다. No
sampling_params object 모델의 샘플링 매개변수입니다. No
type enum 객체 유형은 항상 score_model입니다.
가능한 값: score_model
Yes

EvalStoredCompletionsDataSourceConfig

저장된 완료 쿼리의 메타데이터 속성을 지정하는 StoredCompletionsDataSourceConfig입니다. 보통 이런 메타데이터 usecase=chatbot 는 또는 prompt-version=v2, 이 데이터 소스 구성에서 반환되는 스키마는 평가에서 사용 가능한 변수를 정의하는 데 사용됩니다. itemsample 데이터 소스 설정을 사용할 때 두 가지 모두 정의됩니다.

Name Type Description Required Default
metadata Metadata 객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다. No
schema object 실행 데이터 소스 항목에 대한 json 스키마입니다. 여기에서 JSON 스키마 구축 방법을 배워보세요. Yes
type enum 데이터 소스의 종류. 항상 그렇습니다.stored_completions
가능한 값: stored_completions
Yes

EvalStoredCompletionsSource

필터 집합을 설명하는 StoredCompletionsRunDataSource 구성

Name Type Description Required Default
created_after integer 이 시기 이후에 생성된 항목을 필터링할 수 있는 선택적 유닉스 타임스탬프가 있습니다. No
created_before integer 이 시기 이전에 생성된 항목을 필터링하는 선택적 Unix 타임스탬프가 있습니다. No
limit integer 반품할 수 있는 최대 아이템 수는 선택 사항입니다. No
metadata Metadata 객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다. No
model string 필터링할 수 있는 선택적 모델(예: 'gpt-4o')이 있습니다. No
type enum 출처의 종류. 항상 그렇습니다.stored_completions
가능한 값: stored_completions
Yes

EvalStringCheckGrader

입력과 참조 간 문자열 비교를 수행하는 StringCheckGrader 객체입니다.

Name Type Description Required Default
input string 입력된 텍스트. 여기에는 템플릿 문자열이 포함될 수 있습니다. Yes
name string 채점자의 이름입니다. Yes
operation enum 문자열 검사 연산을 수행하세요. , eqne, like, , 중 ilike하나 중 하나 .
가능한 값들: eq, ne, like, ilike
Yes
reference string 참고 문헌. 여기에는 템플릿 문자열이 포함될 수 있습니다. Yes
type enum 객체 유형은 항상 string_check입니다.
가능한 값: string_check
Yes

EvalTextSimilarityGrader

유사도 지표를 기준으로 텍스트를 평가하는 TextSimilarityGrader 객체입니다.

Name Type Description Required Default
evaluation_metric enum 평가 지표를 사용해야 합니다. , fuzzy_matchbleu, rouge_2meteorgleurouge_3rouge_4rouge_1rouge_5, , . rouge_l중 하나 중 하나가 있습니다.
가능한 값: fuzzy_match, bleu, gleu, meteorrouge_1, rouge_2, rouge_3rouge_4rouge_5, ,rouge_l
Yes
input string 채점하는 텍스트. Yes
name string 채점자의 이름입니다. No
pass_threshold number 부동 점수(float score)가 크거나 큰 경우 합격 점수를 나타냅니다. Yes
reference string 평가되는 텍스트. Yes
type enum 채점자의 종류.
가능한 값: text_similarity
Yes

EventType

콘텐츠 필터 결과의 심각도를 정의합니다.

Property Value
Description 콘텐츠 필터 결과의 심각도를 정의합니다.
Type string
Values message
metrics

File

파일은 교육과 검증에 사용할 수 있는 문서입니다. 또한 결과 세부사항이 포함된 서비스 생성 문서일 수도 있습니다.

Name Type Description Required Default
bytes integer 이 파일의 크기는 제공될 경우(null 상태일 수 있음) 자바스크립트 정수와의 호환성을 보장하기 위해 2^53-1보다 큰 파일 크기는 지원되지 않습니다. No
created_at integer 이 작업이나 항목이 생성된 타임스탬프(유닉스 에포크)입니다. No
expires_at integer 파일이 만료된 유닉스 타임스탬프(1970년 1월 1일부터 경과한 초 수)입니다. No
filename string 파일 이름. Yes
id string 이 물건의 정체. No
object TypeDiscriminator 객체의 유형을 정의합니다. No
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes
status FileState 파일의 상태. No
status_details string 이 파일 처리 실패 시 확인할 수 있는 오류 메시지와 세부 정보를 포함하고 있습니다. Deprecated. No

FileCreate

Azure OpenAI에서 사용할 수 있도록 외부 콘텐츠 URL에서 가져올 문서를 정의합니다.

Name Type Description Required Default
content_url string 문서를 다운로드할 URL은 (블롭의 SAS URL 또는 GET 요청으로 접근 가능한 다른 외부 URL 중 하나일 수 있습니다). Yes
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes

FileDelete

파일 삭제에 대한 응답을 정의합니다.

Name Type Description Required Default
deleted boolean 이 값은 파일이 삭제되었는지 여부를 나타냅니다. Yes
id string 삭제된 파일 아이디. Yes
object string 파일이 삭제되었다면. Yes

FileDetails

파일 디테일은 배치 서비스가 파일 블롭 디테일을 가져오기 위해 사용하는 문서입니다.

Name Type Description Required Default
filename string 파일 이름. Yes
id string 이 물건의 정체. No
path string 컨테이너 내에서 파일에 대한 상대적 경로입니다. No
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes
storage_account_uri string 이 파일의 저장 계정 URL입니다. No
storage_container_name string 이 파일 블롭의 스토리지 컨테이너 이름입니다. No

FileExpiresAfter

파일의 만료일을 정의합니다.

Name Type Description Required Default
anchor FileExpiryAnchor 절대 만료 시점에 대해 앵커를 정의합니다. No
seconds integer 상대 유기량 시간은 초 단위입니다. 범위: [1209600 - 2592000]. No

FileExpiryAnchor

절대 만료 시점에 대해 앵커를 정의합니다.

Property Value
Description 절대 만료 시점에 대해 앵커를 정의합니다.
Type string
Values created_at

FileImport

Azure OpenAI에서 사용할 수 있도록 외부 콘텐츠 URL에서 가져올 문서를 정의합니다.

Name Type Description Required Default
content_url string 문서를 다운로드할 URL은 (블롭의 SAS URL 또는 GET 요청으로 접근 가능한 다른 외부 URL 중 하나일 수 있습니다). Yes
expires_after FileExpiresAfter 파일의 만료일을 정의합니다. No
filename string 업로드할 JSON Lines 파일 이름입니다. 만약 ' purpose 파인튜닝'으로 설정되어 있다면, 각 줄은 'prompt'와 'completion' 필드가 포함된 JSON 레코드로 훈련 예제를 나타냅니다. Yes
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes

FileList

파일 목록을 나타냅니다.

Name Type Description Required Default
data array 물건 목록. No
object TypeDiscriminator 객체의 유형을 정의합니다. No

FileState

파일의 상태.

Property Value
Description 파일의 상태.
Type string
Values uploaded
pending
running
processed
error
deleting
deleted

FineTuneMethod

FineTuneMethod 판별기

이 구성 요소는 다음과 같은 속성을 type 사용하여 서로 다른 유형을 구별합니다:

Type Value Schema
Name Type Description Required Default
type FineTuningMethodType Yes

FineTuneMethodRequest

FineTuneMethodRequest 판별기

이 구성 요소는 다음과 같은 속성을 type 사용하여 서로 다른 유형을 구별합니다:

Type Value Schema
Name Type Description Required Default
type FineTuningMethodType Yes

FineTuningHyperParameters

미세 조정 작업에 사용되는 하이퍼 파라미터 설정들.

Name Type Description Required Default
batch_size integer 학습에 사용할 일괄 처리 크기입니다. 일괄 처리 크기는 단일 앞으로 및 뒤로 패스를 학습하는 데 사용되는 학습 예제의 수입니다. 일반적으로 우리는 더 큰 배치 크기가 더 큰 데이터셋에 더 잘 작동하는 경향이 있음을 발견했습니다. 이 속성의 기본값과 최대 값은 기본 모델에 따라 다릅니다. No
learning_rate_multiplier number 학습에 사용할 학습 속도 승수입니다. 미세 조정 학습률은 사전 학습에 사용된 원래 학습률에 이 값을 곱한 값입니다. 학습 속도가 클수록 일괄 처리 크기가 클수록 성능이 향상되는 경향이 있습니다. 0.02에서 0.2 범위의 값을 실험해 보시고 어떤 값이 가장 좋은 결과를 내는지 확인해 보시길 권장합니다. No
n_epochs integer 모델을 학습시킬 Epoch의 수입니다. Epoch는 학습 데이터 세트를 통해 하나의 전체 주기를 나타냅니다. No

FineTuningJob

미세 조정 작업의 가치를 정의합니다.

Name Type Description Required Default
created_at integer 이 작업이나 항목이 생성된 타임스탬프(유닉스 에포크)입니다. No
error FineTuningJobError 실패한 미세 조정 작업의 경우, 여기에는 실패 원인에 대한 더 많은 정보가 포함됩니다. No
estimated_finish integer 유닉스 타임스탬프(초 단위)는 미세 조정 작업이 완료될 것으로 추정되는 시점을 나타냅니다. 파인튜닝 작업이 실행되지 않으면 값은 null이 됩니다. No
fine_tuned_model string 결과적으로 미세 조정된 모델의 식별자(model-id)입니다. 이 특성은 성공적으로 완료된 미세 조정 실행을 위해 채워집니다. 이 식별자를 사용하여 추론을 위한 배포를 생성하세요. No
finished_at integer 이 작업이나 항목이 성공적으로 완료되었을 때(유닉스 에포크 기준) 타임스탬프입니다. No
hyperparameters FineTuningHyperParameters 미세 조정 작업에 사용되는 하이퍼 파라미터 설정들. No
id string 이 물건의 정체. No
integrations array 미세 조정 작업을 지원하는 통합 구성 목록입니다. 여러 통합 작업이 계획되어 있으니, 통합 유형을 확인하고 행동하는 것이 중요합니다. No
method object 감독 하 미세 조정에 사용되는 방법 No
└─ type FineTuningMethodType No
model string 미세 조정에 사용된 기본 모델의 식별자(model-id). Yes
object TypeDiscriminator 객체의 유형을 정의합니다. No
organisation_id string 이 미세 조정 작업의 조직 ID. Azure OpenAI에서는 사용하지 않음; OpenAI 호환성만 지원합니다. No
result_files array 결과 파일 식별자(file-id)는 csv 형식으로 훈련 및 평가 지표를 포함합니다. 이 파일은 성공적으로 완료된 미세 조정 실행에만 제공됩니다. No
seed integer 미세 조정에 사용되는 씨앗. No
status FineTuningState 미세 조정 객체 또는 미세 조정 작업의 상태입니다. No
suffix string 미세 조정 모델을 식별하는 데 사용된 접미사. 접미사는 최대 40자(a-z, A-Z, 0-9, _)까지 포함할 수 있으며, 이는 미세 조정된 모델명에 추가됩니다. No
trained_tokens integer 이 미세 조정 작업으로 처리된 총 청구 가능한 토큰 수입니다. No
training_file string 훈련에 사용되는 파일입니다. Yes
validation_file string 이 파일은 훈련 중 미세 조정된 모델을 평가하는 데 사용됩니다. No

FineTuningJobCheckpoint

Name Type Description Required Default
created_at integer 체크포인트가 생성된 시점의 유닉스 타임스탬프(초 단위)입니다. Yes
fine_tuned_model_checkpoint string 세밀하게 조정된 체크포인트 모델의 이름입니다. Yes
fine_tuning_job_id string 이 체크포인트가 만들어진 미세 조정 작업의 이름입니다. Yes
id string 이 검문소의 정체. No
metrics object 미세 조정 작업 중 단계에 있는 지표들. Yes
object TypeDiscriminator 객체의 유형을 정의합니다. No
step_number integer 체크포인트가 생성된 단계 번호입니다. Yes

FineTuningJobCheckpointList

체크포인트 목록을 나타냅니다.

Name Type Description Required Default
data array 물건 목록. No
has_more boolean 리스트에 반환된 것보다 더 많은 요소를 포함하고 있는지를 나타내는 값입니다. No
object TypeDiscriminator 객체의 유형을 정의합니다. No

FineTuningJobCreation

미세 조정 작업의 가치를 정의합니다.

Name Type Description Required Default
hyperparameters FineTuningHyperParameters 미세 조정 작업에 사용되는 하이퍼 파라미터 설정들. No
integrations array 미세 조정 작업을 지원하는 통합 구성 목록입니다. 여러 통합 작업이 계획되어 있으니, 통합 유형을 확인하고 행동하는 것이 중요합니다. No
method object 감독 하 미세 조정에 사용되는 방법 No
└─ type FineTuningMethodType No
model string 이 미세 조정에 사용되는 기본 모델의 식별자(model-id)입니다. Yes
seed integer 미세 조정에 사용되는 씨앗. 동일한 시드와 작업 매개변수를 전달해도 결과는 동일하지만, 드물게 차이가 있을 수 있습니다. 시드가 지정되지 않았다면, 하나가 생성되어 줍니다. No
suffix string 미세 조정 모델을 식별하는 데 사용된 접미사. 접미사는 최대 40자(a-z, A-Z, 0-9, _)까지 포함할 수 있으며, 이는 미세 조정된 모델명에 추가됩니다. No
training_file string 이 미세 조정 모델을 학습하는 데 사용되는 파일 아이덴티티(file-id)입니다. Yes
validation_file string 훈련 중 미세 조정된 모델을 평가하는 데 사용되는 파일 식별자(file-id). No

FineTuningJobError

실패한 미세 조정 작업의 경우, 여기에는 실패 원인에 대한 더 많은 정보가 포함됩니다.

Name Type Description Required Default
code string 기계 읽기 가능한 오류 코드. No
message string 사람이 읽을 수 있는 오류 메시지. No
param string 무효인 매개변수, 보통 training_file 또는 validation_file. 실패가 매개변수 특이화가 아니라면 이 필드는 영(null)이 됩니다. No

FineTuningJobEvent

Name Type Description Required Default
created_at integer 이 이벤트가 생성된 타임스탬프(유닉스 에포흐 기준)입니다. Yes
data 이 사건의 기계 판독 가능한 데이터. No
id string 이 사건의 정체. No
level LogLevel 행사의 장황한 수준. Yes
message string 그 사건을 설명하는 메시지. 이는 상태 변경(예: 대기열, 시작, 실패, 완료)일 수도 있고, 업로드된 결과 같은 다른 이벤트일 수 있습니다. Yes
object TypeDiscriminator 객체의 유형을 정의합니다. No
type EventType 콘텐츠 필터 결과의 심각도를 정의합니다. Yes

FineTuningJobEventList

사건 목록을 나타냅니다.

Name Type Description Required Default
data array 물건 목록. No
has_more boolean 리스트에 반환된 것보다 더 많은 요소를 포함하고 있는지를 나타내는 값입니다. No
object TypeDiscriminator 객체의 유형을 정의합니다. No

FineTuningJobList

이것은 미세한 곡들의 목록을 나타냅니다.

Name Type Description Required Default
data array 물건 목록. No
has_more boolean 리스트에 반환된 것보다 더 많은 요소를 포함하고 있는지를 나타내는 값입니다. No
object TypeDiscriminator 객체의 유형을 정의합니다. No

FineTuningMethodType

Property Value
Type string
Values dpo
supervised

FineTuningState

미세 조정 객체 또는 미세 조정 작업의 상태입니다.

Property Value
Description 미세 조정 객체 또는 미세 조정 작업의 상태입니다.
Type string
Values created
pending
running
pausing
paused
resuming
succeeded
cancelled
failed

FunctionCallFunction

Name Type Description Required Default
arguments string 모델이 함수에 전달할 것으로 기대하는 인수들입니다. No
name string 함수 이름입니다. No

FunctionDefinition

Name Type Description Required Default
description string 함수가 수행하는 기능에 대한 설명으로, 모델이 언제 어떻게 함수를 호출할지 결정하는 데 사용됩니다. No
name string 호출할 함수의 이름입니다. No
parameters 함수들이 받아들이는 매개변수는 JSON 스키마 객체로 설명됩니다. No

GenericEmbeddingSettings

연결 임베딩 설정

Name Type Description Required Default
connection BaseConnection 자원과의 연결고리. No
deploymentName string No
modelName string No

IngestionError

섭취 오류의 세부 사항

Name Type Description Required Default
innerErrors array No
message string No

IngestionJob

일의 세부 사항을 나타냅니다.

섭취 분류자 Job에 대한

이 구성 요소는 다음과 같은 속성을 kind 사용하여 서로 다른 유형을 구별합니다:

Type Value Schema
Name Type Description Required Default
jobId string No
kind IngestionJobType 직업 유형. Yes

IngestionJobList

섭취 작업 목록을 나타냅니다.

Name Type Description Required Default
nextLink string 다음 페이지로 가는 링크입니다. No
value array 물건 목록. No

IngestionJobProgress

Name Type Description Required Default
stageProgress array Yes

IngestionJobRun

작업 진행의 세부 사항

Name Type Description Required Default
error IngestionError 섭취 오류의 세부 사항 No
jobId string No
progress IngestionJobProgress No
runId string No
status OperationState 작업이나 아이템의 상태. No
warnings array No

IngestionJobRunList

이는 인제스티션 작업 실행 목록을 나타냅니다.

Name Type Description Required Default
nextLink string 다음 페이지로 가는 링크입니다. No
value array 물건 목록. No

IngestionJobStageProgress

Name Type Description Required Default
name string No
processedItems integer No
state OperationState 작업이나 아이템의 상태. No
subStageProgress array No
totalItems integer No

IngestionJobSystemCompute

Name Type Description Required Default
completionAction CompletionAction 완성 동작. No
dataRefreshIntervalInHours integer No
datasource SystemComputeDatasource No
jobId string No
kind IngestionJobType 직업 유형. Yes
searchServiceConnection BaseConnection 자원과의 연결고리. No

IngestionJobType

직업 유형.

Property Value
Description 직업 유형.
Type string
Values SystemCompute
UserCompute

IngestionJobUserCompute

Name Type Description Required Default
compute JobCompute 작업의 컴퓨트 설정입니다. No
dataRefreshIntervalInHours integer No
datasource UserComputeDatasource No
jobId string No
kind IngestionJobType 직업 유형. Yes
target TargetIndex 작성할 인덱스에 관한 정보. No
workspaceId string No

InnerError

내부 오류는 Microsoft REST 가이드라인에 정의된 것입니다.

Name Type Description Required Default
code InnerErrorCode 내부 오류 코드는 Microsoft REST 가이드라인에 정의되어 있습니다. No
innererror InnerError 내부 오류는 Microsoft REST 가이드라인에 정의된 것입니다. No

InnerErrorCode

내부 오류 코드는 Microsoft REST 가이드라인에 정의되어 있습니다.

Property Value
Description 내부 오류 코드는 Microsoft REST 가이드라인에 정의되어 있습니다.
Type string
Values invalidPayload

InputMessage

Name Type Description Required Default
content string 메시지 내용. No
id string 메시지의 정체성. No
name string 참가자의 선택 이름입니다. 동일한 역할의 참가자들을 구분할 수 있는 모델 정보를 제공합니다. No
role string 메시지를 생성하는 주체의 역할. No

InputTextContent

모델에 입력하는 텍스트.

Name Type Description Required Default
text string 모델에 입력하는 텍스트. Yes
type enum 입력 항목의 종류. 항상 그렇습니다.input_text
가능한 값: input_text
Yes

IntegrationTypes

사용 가능한 미세 조정 통합 방법 목록.

Property Value
Description 사용 가능한 미세 조정 통합 방법 목록.
Type string
Values wandb

JobCompute

작업의 컴퓨트 설정입니다.

JobCompute용 판별기

이 구성 요소는 다음과 같은 속성을 kind 사용하여 서로 다른 유형을 구별합니다:

Type Value Schema
Name Type Description Required Default
kind ComputeType 컴퓨트 타입입니다. Yes

LifeCycleStatus

모델의 수명 주기 상태.

참고: 모델은 에서 previewgenerally-available에서 로 승격할 수 있지만, 에서 generally-availablepreview에서 로 승격할 수는 없습니다.

Property Value
Description 모델의 수명 주기 상태. 참고: 모델은 에서 previewgenerally-available에서 로 승격할 수 있지만, 에서 generally-availablepreview에서 로 승격할 수는 없습니다.
Type string
Values preview
generally-available
deprecating
deprecated

LogLevel

행사의 장황한 수준.

Property Value
Description 행사의 장황한 수준.
Type string
Values info
warning
error

Logprob

Name Type Description Required Default
logprob number 이 토큰이 상위 20개 토큰 안에 들 경우 로그 확률입니다. 그렇지 않으면 이 값 -9999.0 은 토큰이 매우 희박하다는 의미로 사용됩니다. No
token string The token. No

Metadata

객체에 부착할 수 있는 16개의 키값 쌍 집합입니다. 이는 객체에 대한 추가 정보를 구조화된 형식으로 저장하고, API나 대시보드를 통해 객체를 쿼리하는 데 유용할 수 있습니다. 키는 최대 64자 길이의 문자열입니다. 값은 최대 512자 길이의 문자열입니다.

이 컴포넌트에 대한 속성이 정의되어 있지 않습니다.

Model

모델은 기본 모델이거나 성공적인 미세 조정 작업의 결과물입니다.

Name Type Description Required Default
capabilities Capabilities 기본 모델이나 미세 조정 모델의 기능. Yes
created_at integer 이 작업이나 항목이 생성된 타임스탬프(유닉스 에포크)입니다. No
deprecation Deprecation 모델의 다양한 사용 사례에 대한 폐기일을 정의합니다. 보통 기본 모델은 생성 후 1년간의 미세 조정을 지원하며, 기본 모델이나 미세 조정 모델 생성 후 2년이 지난 후에도 보통 지원됩니다. 정확한 날짜는 부동산에 명시되어 있습니다. Yes
fine_tune string 미세 조정 작업 식별자(fine-tune-id)는 미세 조정 모델이라면; 그렇지 않으면 null. No
id string 이 물건의 정체. No
lifecycle_status LifeCycleStatus 모델의 수명 주기 상태. 참고: 모델은 "preview"에서 로 승격할 수 있지만, "previewgenerally-available"generally-available로 승격할 수는 없습니다. Yes
model string 미세 조정 모델이라면 기본 모델 정체성(model-id); 그렇지 않으면 null. No
object TypeDiscriminator 객체의 유형을 정의합니다. No
status FineTuningState 미세 조정 객체 또는 미세 조정 작업의 상태입니다. No

ModelList

모델 목록을 나타냅니다.

Name Type Description Required Default
data array 물건 목록. No
object TypeDiscriminator 객체의 유형을 정의합니다. No

OperationState

작업이나 아이템의 상태.

Property Value
Description 작업이나 아이템의 상태.
Type string
Values notRunning
running
succeeded
failed

Order

파일의 목적을 정의합니다.

Property Value
Description 파일의 목적을 정의합니다.
Type string
Values desc
asc

OrderByOrder

Property Value
Type string
Values asc
desc

PineconeIndex

Pinecone Index.

Name Type Description Required Default
connectionId string 연결 ID가 솔방울을 가리키고 있습니다. No
kind TargetType 타깃 유형. Yes

Purpose

업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다.

Property Value
Description 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다.
Type string
Values fine-tune
fine-tune-results
assistants
assistants_output
batch
batch_output
evals
evals_output

ServerlessCompute

Serverless compute.

Name Type Description Required Default
instanceCount integer 작업을 실행할 인스턴스 수. No
kind ComputeType 컴퓨트 타입입니다. Yes
sku string SKU Level No

StoredCompletion

Name Type Description Required Default
choices array 채팅 완료 선택 목록입니다. 가 1보다 크면 n 1 이상일 수 있습니다. No
created integer 채팅 완료 시점의 유닉스 타임스탬프(초 단위)입니다. No
frequency_penalty number 수치는 -2.0에서 2.0 사이입니다. 양수(positive)는 지금까지 텍스트 내 기존 빈도에 따라 새로운 토큰에 페널티를 적용하여, 같은 줄을 그대로 반복할 가능성을 줄입니다. No
id string 저장된 완성의 동일식. No
input_user string 이 요청의 입력 사용자입니다. No
metadata object 추가 정보를 위한 임의의 키-값 쌍. No
model string 사용할 모델의 ID. No
presence_penalty number 수치는 -2.0에서 2.0 사이입니다. 양의 값은 지금까지 텍스트에 등장했는지 여부에 따라 새로운 토큰에 페널티를 적용하여, 모델이 새로운 주제를 다룰 가능성을 높입니다. No
request_id string OpenAI API 요청에 대한 고유 식별자입니다. 지원팀에 연락할 때 이 요청 ID를 꼭 포함해 주세요. No
seed integer 명시가 명시되어 있을 경우, 우리 시스템은 동일한 seed 매개변수로 반복 요청이 같은 결과를 반환하도록 결정론적으로 샘플링하려 최선을 다합니다. 결정론이 보장되는 것은 아니며, 참고하시는 것이 좋습니다
system_fingerprint 백엔드 변경 사항을 모니터링하기 위한 응답 매개변수. No
service_tier string 요청을 처리하는 데 사용할 지연 계층을 지정합니다. No
system_fingerprint string 이 지문은 모델이 실행하는 백엔드 구성을 나타냅니다. 요청 매개변수와 seed 함께 사용하여 백엔드 변경이 이루어져 결정론에 영향을 줄 수 있음을 이해할 수 있습니다. No
temperature number 샘플링 온도는 0에서 2 사이로 사용해야 합니다. 0.8처럼 높은 값은 출력이 더 무작위적이고, 0.2처럼 낮은 값은 더 집중되고 결정론적으로 만듭니다. 일반적으로 이 부분을 변경하거나 top_p 둘 다 바꾸는 것을 권장하지 않습니다. No
tool_choice string 모델이 호출하는 도구가 어떤 것인지(있다면) 제어합니다. No
tools array 모델이 호출할 수 있는 도구 목록입니다. 현재는 함수만 지원되고 있습니다. 이를 사용해 모델이 JSON 입력을 생성할 수 있는 함수 목록을 제공하세요. 최대 128개의 기능을 지원합니다. No
top_p number 온도에 따른 샘플링의 대안인 핵 샘플링으로, 모델이 확률 질량을 가진 토큰의 결과를 고려하는 경우top_p. 따라서 0.1은 상위 10개의% 확률 질량을 구성하는 토큰만 고려된다는 뜻입니다. 일반적으로 이 부분을 변경하거나 temperature 둘 다 바꾸는 것을 권장하지 않습니다. No
usage Usage No

StoredCompletionDelete

Name Type Description Required Default
deleted boolean 저장된 완료가 삭제되었는지를 나타내는 값입니다. No
id string 삭제된 저장된 완료 ID입니다. No
object string 저장된 완료가 삭제되었다면. Yes

StoredCompletionList

Name Type Description Required Default
data array 회수된 물건 목록. No
first_id string 검색된 첫 번째 ID list No
has_more boolean 이 부동산은 has_more 추가 결과가 있음을 나타내기 위해 페이지네이션에 사용됩니다. No
last_id string 마지막으로 검색된 신원 list No
object string 객체 유형은 항상 list No
total integer 총 아이템 수. No

StoredCompletionMessages

Name Type Description Required Default
data array 회수된 물건 목록. No
first_id string 검색된 첫 번째 ID list No
has_more boolean 이 부동산은 has_more 추가 결과가 있음을 나타내기 위해 페이지네이션에 사용됩니다. No
last_id string 마지막으로 검색된 신원 list No
object string 객체 유형은 항상 list No
total integer 총 아이템 수. No

StoredCompletionResponse

Name Type Description Required Default
choices array 채팅 완료 선택 목록입니다. 가 1보다 크면 n 1 이상일 수 있습니다. No
created integer 채팅 완료 시점의 유닉스 타임스탬프(초 단위)입니다. No
frequency_penalty number 수치는 -2.0에서 2.0 사이입니다. 양수(positive)는 지금까지 텍스트 내 기존 빈도에 따라 새로운 토큰에 페널티를 적용하여, 같은 줄을 그대로 반복할 가능성을 줄입니다. No
id string 저장된 완성의 동일식. No
input_user string 이 요청의 입력 사용자입니다. No
metadata object 추가 정보를 위한 임의의 키-값 쌍. No
model string 사용할 모델의 ID. No
object string 이 물체의 종류. No
presence_penalty number 수치는 -2.0에서 2.0 사이입니다. 양의 값은 지금까지 텍스트에 등장했는지 여부에 따라 새로운 토큰에 페널티를 적용하여, 모델이 새로운 주제를 다룰 가능성을 높입니다. No
request_id string OpenAI API 요청에 대한 고유 식별자입니다. 지원팀에 연락할 때 이 요청 ID를 꼭 포함해 주세요. No
seed integer 명시가 명시되어 있을 경우, 우리 시스템은 동일한 seed매개변수로 반복 요청이 같은 결과를 반환하도록 결정론적으로 샘플링하려 최선을 다합니다. 결정론이 보장되는 것은 아니므로, 백엔드 변경 사항을 모니터링하려면 응답 매개변수를 참고 system_fingerprint 해야 합니다. No
service_tier string 요청을 처리하는 데 사용할 지연 계층을 지정합니다. No
system_fingerprint string 이 지문은 모델이 실행하는 백엔드 구성을 나타냅니다. 요청 매개변수와 seed 함께 사용하여 백엔드 변경이 이루어져 결정론에 영향을 줄 수 있음을 이해할 수 있습니다. No
temperature number 샘플링 온도는 0에서 2 사이로 사용해야 합니다. 0.8처럼 높은 값은 출력이 더 무작위적이고, 0.2처럼 낮은 값은 더 집중되고 결정론적으로 만듭니다. 일반적으로 이 부분을 변경하거나 top_p 둘 다 바꾸는 것을 권장하지 않습니다. No
tool_choice string 모델이 호출하는 도구가 어떤 것인지(있다면) 제어합니다. No
tools array 모델이 호출할 수 있는 도구 목록입니다. 현재는 함수만 지원되고 있습니다. 이를 사용해 모델이 JSON 입력을 생성할 수 있는 함수 목록을 제공하세요. 최대 128개의 기능을 지원합니다. No
top_p number 온도에 따른 샘플링의 대안인 핵 샘플링으로, 모델이 확률 질량을 가진 토큰의 결과를 고려하는 경우top_p. 따라서 0.1은 상위 10개의% 확률 질량을 구성하는 토큰만 고려된다는 뜻입니다. 일반적으로 이 부분을 변경하거나 temperature 둘 다 바꾸는 것을 권장하지 않습니다. No
usage Usage No

SupervisedMethod

감독 하 미세 조정에 사용되는 방법

Name Type Description Required Default
type FineTuningMethodType Yes

SupervisedMethodRequest

감독 하 미세 조정에 사용되는 방법

Name Type Description Required Default
type FineTuningMethodType Yes

SystemComputeDatasource

SystemComputeDatasource용 판별기

이 구성 요소는 다음과 같은 속성을 kind 사용하여 서로 다른 유형을 구별합니다:

Type Value Schema
Name Type Description Required Default
kind SystemComputeDatasourceType 데이터소스 유형. Yes

SystemComputeDatasourceType

데이터소스 유형.

Property Value
Description 데이터소스 유형.
Type string
Values Storage
Urls

SystemComputeStorage

Storage account

Name Type Description Required Default
chunking ChunkingSettings Chunking settings No
connection BaseConnection 자원과의 연결고리. No
containerName string container name No
embeddings array No
kind SystemComputeDatasourceType 데이터소스 유형. Yes

SystemComputeUrl

Urls

Name Type Description Required Default
chunking ChunkingSettings Chunking settings No
connection BaseConnection 자원과의 연결고리. No
containerName string container name No
crawling CrawlingSettings Crawling settings No
embeddings array No
kind SystemComputeDatasourceType 데이터소스 유형. Yes
urls array No

TargetIndex

작성할 인덱스에 관한 정보.

TargetIndex 판별기

이 구성 요소는 다음과 같은 속성을 kind 사용하여 서로 다른 유형을 구별합니다:

Type Value Schema
Name Type Description Required Default
kind TargetType 타깃 유형. Yes

TargetType

타깃 유형.

Property Value
Description 타깃 유형.
Type string
Values AzureAISearch
CosmosDB
Pinecone

Tool

Name Type Description Required Default
function FunctionDefinition No
type string 툴 콜의 종류. 이런 function 종류의 공구 호출을 위한 것입니다. No

ToolCall

Name Type Description Required Default
function FunctionCallFunction No
id string 툴 호출의 ID입니다. No
type string 출력이 요구되는 툴 호출 유형. 지금은 항상 function. No

TypeDiscriminator

객체의 유형을 정의합니다.

Property Value
Description 객체의 유형을 정의합니다.
Type string
Values list
fine_tuning.job
file
fine_tuning.job.event
fine_tuning.job.checkpoint
model
upload
upload.part
batch
wandb

UpdateStoredCompletionRequest

Name Type Description Required Default
metadata object 추가 정보를 위한 임의의 키-값 쌍. No

UploadFileCompleteBody

Name Type Description Required Default
md5 string No
part_ids array Yes

UploadFileStartBody

Name Type Description Required Default
bytes integer Yes
filename string Yes
mime_type string Yes
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. Yes

UploadPartResource

Name Type Description Required Default
azure_block_id string No
created_at integer 이 작업이나 항목이 생성된 타임스탬프(유닉스 에포크)입니다. No
id string 이 물건의 정체. No
object TypeDiscriminator 객체의 유형을 정의합니다. No
upload_id string No

UploadResource

Name Type Description Required Default
bytes integer No
created_at integer 이 작업이나 항목이 생성된 타임스탬프(유닉스 에포크)입니다. No
expires_at integer No
file File 파일은 교육과 검증에 사용할 수 있는 문서입니다. 또한 결과 세부사항이 포함된 서비스 생성 문서일 수도 있습니다. No
filename string No
id string 이 물건의 정체. No
object TypeDiscriminator 객체의 유형을 정의합니다. No
purpose Purpose 업로드된 문서의 의도된 목적. 미세 조정에는 '파인-튜닝'을 사용하세요. 이를 통해 업로드된 파일의 형식을 검증할 수 있습니다. No
status UploadStatus No

UploadStatus

Property Value
Type string
Values pending
expired
completed
cancelled

Usage

Name Type Description Required Default
completion_tokens integer 생성된 완료 토큰의 수. No
prompt_tokens integer 프롬프트에 있는 토큰 수. No
total_tokens integer 요청에 사용된 총 토큰 수(프롬프트 + 완료). No

UserComputeDataset

Storage account

Name Type Description Required Default
chunking ChunkingSettings Chunking settings No
datasetId string No
datasetType string No
embeddings array No
kind UserComputeDatasourceType 데이터소스 유형. Yes

UserComputeDatasource

UserComputeDatasource용 판별기

이 구성 요소는 다음과 같은 속성을 kind 사용하여 서로 다른 유형을 구별합니다:

Type Value Schema
Name Type Description Required Default
kind UserComputeDatasourceType 데이터소스 유형. Yes

UserComputeDatasourceType

데이터소스 유형.

Property Value
Description 데이터소스 유형.
Type string
Values Dataset
Urls

UserComputeUrl

Urls

Name Type Description Required Default
chunking ChunkingSettings Chunking settings No
crawling CrawlingSettings Crawling settings No
embeddings array No
kind UserComputeDatasourceType 데이터소스 유형. Yes
urls array No

WandBIntegrationRequest

Name Type Description Required Default
entity string No
name string No
project string Yes
tags array No

WandBIntegrationRequestWrapper

Name Type Description Required Default
type IntegrationTypes 사용 가능한 미세 조정 통합 목록. Yes
wandb WandBIntegrationRequest Yes

WorkspaceConnection

AML 작업 공간 연결.

Name Type Description Required Default
connectionId string ConnectionId No
kind ConnectionType 연결 유형. Yes

WorkspaceConnectionEmbeddingSettings

임베딩 모델과의 연결 ID

Name Type Description Required Default
connectionId string No
deploymentName string No
modelName string No

Next steps

모델에 대해 배우 고, REST API를 활용한 미세 조정도 알아보세요. OpenAI를 구동하는 기본 모델에 대해 더 알아보세요.