Azure OpenAI 추론 모델

Azure OpenAI 추론 모델은 더욱 향상된 집중력과 기능으로 추론 및 문제 해결 작업을 처리하도록 설계되었습니다. 이러한 모델은 사용자의 요청을 처리하고 이해하는 데 더 많은 시간을 할애하여 이전 반복에 비해 과학, 코딩 및 수학과 같은 영역에서 매우 강력합니다.

추론 모델의 주요 기능:

• 복잡한 코드 생성: 개발자를 지원하기 위해 알고리즘을 생성하고 고급 코딩 작업을 처리할 수 있습니다.
• 고급 문제 해결: 포괄적인 브레인스토밍 세션 및 다각적인 문제 해결에 이상적입니다.
• 복잡한 문서 비교: 계약, 사례 파일 또는 법적 문서를 분석하여 미묘한 차이를 식별하는 데 적합합니다.
• 지침 따르기 및 워크플로 관리: 더 짧은 컨텍스트가 필요한 워크플로를 관리하는 데 특히 효과적입니다.

가용도

지역 가용성

API 및 기능 지원

지원되지 않음

다음은 현재 추론 모델에서 지원되지 않습니다.

• temperature, top_p, presence_penalty, frequency_penalty, logprobs, top_logprobs, logit_biasmax_tokens

사용법

이러한 모델은 현재 채팅 완료 API를 사용하는 다른 모델과 동일한 매개 변수 집합을 지원하지 않습니다 .

pip install openai --upgrade

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("AZURE_OPENAI_API_KEY"),
    base_url="https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/",
)

response = client.chat.completions.create(
    model="gpt-5-mini", # replace with the model deployment name of your o1 deployment.
    messages=[
        {"role": "user", "content": "What steps should I think about when writing my first Python API?"},
    ],
    max_completion_tokens = 5000

)

print(response.model_dump_json(indent=2))

pip install openai --upgrade

from openai import AzureOpenAI
from azure.identity import DefaultAzureCredential, get_bearer_token_provider

token_provider = get_bearer_token_provider(
    DefaultAzureCredential(), "https://cognitiveservices.azure.com/.default"
)

client = AzureOpenAI(
  azure_endpoint = os.getenv("AZURE_OPENAI_ENDPOINT"), 
  azure_ad_token_provider=token_provider,
  api_version="2025-04-01-preview"
)

response = client.chat.completions.create(
    model="o1-new", # replace with your model deployment name 
    messages=[
        {"role": "user", "content": "What steps should I think about when writing my first Python API?"},
    ],
    max_completion_tokens = 5000

)

print(response.model_dump_json(indent=2))

using Azure.Identity;
using OpenAI;
using OpenAI.Chat;
using System.ClientModel.Primitives;

#pragma warning disable OPENAI001 //currently required for token based authentication

BearerTokenPolicy tokenPolicy = new(
    new DefaultAzureCredential(),
    "https://cognitiveservices.azure.com/.default");

ChatClient client = new(
    model: "o4-mini",
    authenticationPolicy: tokenPolicy,
    options: new OpenAIClientOptions()
    {

        Endpoint = new Uri("https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1")
    }
);

ChatCompletionOptions options = new ChatCompletionOptions
{
    MaxOutputTokenCount = 100000
};

ChatCompletion completion = client.CompleteChat(
         new DeveloperChatMessage("You are a helpful assistant"),
         new UserChatMessage("Tell me about the bitter lesson")
    );

Console.WriteLine($"[ASSISTANT]: {completion.Content[0].Text}");

Python 출력:

{
  "id": "chatcmpl-AEj7pKFoiTqDPHuxOcirA9KIvf3yz",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "logprobs": null,
      "message": {
        "content": "Writing your first Python API is an exciting step in developing software that can communicate with other applications. An API (Application Programming Interface) allows different software systems to interact with each other, enabling data exchange and functionality sharing. Here are the steps you should consider when creating your first Python API...truncated for brevity.",
        "refusal": null,
        "role": "assistant",
        "function_call": null,
        "tool_calls": null
      },
      "content_filter_results": {
        "hate": {
          "filtered": false,
          "severity": "safe"
        },
        "protected_material_code": {
          "filtered": false,
          "detected": false
        },
        "protected_material_text": {
          "filtered": false,
          "detected": false
        },
        "self_harm": {
          "filtered": false,
          "severity": "safe"
        },
        "sexual": {
          "filtered": false,
          "severity": "safe"
        },
        "violence": {
          "filtered": false,
          "severity": "safe"
        }
      }
    }
  ],
  "created": 1728073417,
  "model": "o1-2024-12-17",
  "object": "chat.completion",
  "service_tier": null,
  "system_fingerprint": "fp_503a95a7d8",
  "usage": {
    "completion_tokens": 1843,
    "prompt_tokens": 20,
    "total_tokens": 1863,
    "completion_tokens_details": {
      "audio_tokens": null,
      "reasoning_tokens": 448
    },
    "prompt_tokens_details": {
      "audio_tokens": null,
      "cached_tokens": 0
    }
  },
  "prompt_filter_results": [
    {
      "prompt_index": 0,
      "content_filter_results": {
        "custom_blocklists": {
          "filtered": false
        },
        "hate": {
          "filtered": false,
          "severity": "safe"
        },
        "jailbreak": {
          "filtered": false,
          "detected": false
        },
        "self_harm": {
          "filtered": false,
          "severity": "safe"
        },
        "sexual": {
          "filtered": false,
          "severity": "safe"
        },
        "violence": {
          "filtered": false,
          "severity": "safe"
        }
      }
    }
  ]
}

추론 노력

개발자 메시지

기능적으로 개발자 메시지는 "role": "developer" 시스템 메시지와 동일합니다.

이전 코드 예제에 개발자 메시지를 추가하면 다음과 같이 표시됩니다.

pip install openai --upgrade

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("AZURE_OPENAI_API_KEY"),
    base_url="https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/",
)

response = client.chat.completions.create(
    model="gpt-5-mini", # replace with the model deployment name of your o1 deployment.
    messages=[
        {"role": "developer","content": "You are a helpful assistant."}, # optional equivalent to a system message for reasoning models 
        {"role": "user", "content": "What steps should I think about when writing my first Python API?"},
    ],
    max_completion_tokens = 5000,
    reasoning_effort = "medium" # low, medium, or high
)

print(response.model_dump_json(indent=2))

pip install openai --upgrade

from openai import AzureOpenAI
from azure.identity import DefaultAzureCredential, get_bearer_token_provider

token_provider = get_bearer_token_provider(
    DefaultAzureCredential(), "https://cognitiveservices.azure.com/.default"
)

client = AzureOpenAI(
  azure_endpoint = os.getenv("AZURE_OPENAI_ENDPOINT"), 
  azure_ad_token_provider=token_provider,
  api_version="2025-04-01-preview"
)

response = client.chat.completions.create(
    model="o1-new", # replace with your model deployment name 
    messages=[
        {"role": "developer","content": "You are a helpful assistant."}, # optional equivalent to a system message for reasoning models 
        {"role": "user", "content": "What steps should I think about when writing my first Python API?"},
    ],
    max_completion_tokens = 5000,
    reasoning_effort = "medium" # low, medium, or high

)

print(response.model_dump_json(indent=2))

using Azure.Identity;
using OpenAI;
using OpenAI.Chat;
using System.ClientModel.Primitives;

#pragma warning disable OPENAI001 //currently required for token based authentication

BearerTokenPolicy tokenPolicy = new(
    new DefaultAzureCredential(),
    "https://cognitiveservices.azure.com/.default");

ChatClient client = new(
    model: "o4-mini",
    authenticationPolicy: tokenPolicy,
    options: new OpenAIClientOptions()
    {

        Endpoint = new Uri("https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1")
    }
);

ChatCompletionOptions options = new ChatCompletionOptions
{
    ReasoningEffortLevel = ChatReasoningEffortLevel.Low,
    MaxOutputTokenCount = 100000
};

ChatCompletion completion = client.CompleteChat(
         new DeveloperChatMessage("You are a helpful assistant"),
         new UserChatMessage("Tell me about the bitter lesson")
    );

Console.WriteLine($"[ASSISTANT]: {completion.Content[0].Text}");

추론 요약

응답 API와 함께 최신 추론 모델을 사용하는 경우 추론 요약 매개 변수를 사용하여 모델의 생각의 사슬 추론에 대한 요약을 받을 수 있습니다.

pip install openai --upgrade

import os
from openai import OpenAI

client = OpenAI(  
  base_url = "https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/",
  api_key=os.getenv("AZURE_OPENAI_API_KEY")  
)

response = client.responses.create(
    input="Tell me about the curious case of neural text degeneration",
    model="gpt-5", # replace with model deployment name
    reasoning={
        "effort": "medium",
        "summary": "auto" # auto, concise, or detailed, gpt-5 series do not support concise 
    },
    text={
        "verbosity": "low" # New with GPT-5 models
    }
)

print(response.model_dump_json(indent=2))

curl -X POST "https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/responses" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AZURE_OPENAI_AUTH_TOKEN" \
 -d '{
     "model": "gpt-5",
     "input": "Tell me about the curious case of neural text degeneration",
     "reasoning": {"summary": "auto"},
     "text": {"verbosity": "low"}
    }'

{
  "id": "resp_689a0a3090808190b418acf12b5cc40e0fc1c31bc69d8719",
  "created_at": 1754925616.0,
  "error": null,
  "incomplete_details": null,
  "instructions": null,
  "metadata": {},
  "model": "gpt-5",
  "object": "response",
  "output": [
    {
      "id": "rs_689a0a329298819095d90c34dc9b80db0fc1c31bc69d8719",
      "summary": [],
      "type": "reasoning",
      "encrypted_content": null,
      "status": null
    },
    {
      "id": "msg_689a0a33009881909fe0fcf57cba30200fc1c31bc69d8719",
      "content": [
        {
          "annotations": [],
          "text": "Neural text degeneration refers to the ways language models produce low-quality, repetitive, or vacuous text, especially when generating long outputs. It’s “curious” because models trained to imitate fluent text can still spiral into unnatural patterns. Key aspects:\n\n- Repetition and loops: The model repeats phrases or sentences (“I’m sorry, but...”), often due to high-confidence tokens reinforcing themselves.\n- Loss of specificity: Vague, generic, agreeable text that avoids concrete details.\n- Drift and contradiction: The output gradually departs from context or contradicts itself over long spans.\n- Exposure bias: During training, models see gold-standard prefixes; at inference, they must condition on their own imperfect outputs, compounding errors.\n- Likelihood vs. quality mismatch: Maximizing token-level likelihood doesn’t align with human preferences for diversity, coherence, or factuality.\n- Token over-optimization: Frequent, safe tokens get overused; certain phrases become attractors.\n- Entropy collapse: With greedy or low-temperature decoding, the distribution narrows too much, causing repetitive, low-entropy text.\n- Length and beam search issues: Larger beams or long generations can favor bland, repetitive sequences (the “likelihood trap”).\n\nCommon mitigations:\n\n- Decoding strategies:\n  - Top-k, nucleus (top-p), or temperature sampling to keep sufficient entropy.\n  - Typical sampling and locally typical sampling to avoid dull but high-probability tokens.\n  - Repetition penalties, presence/frequency penalties, no-repeat n-grams.\n  - Contrastive decoding (and variants like DoLa) to filter generic continuations.\n  - Min/max length, stop sequences, and beam search with diversity/penalties.\n\n- Training and alignment:\n  - RLHF/DPO to better match human preferences for non-repetitive, helpful text.\n  - Supervised fine-tuning on high-quality, diverse data; instruction tuning.\n  - Debiasing objectives (unlikelihood training) to penalize repetition and banned patterns.\n  - Mixture-of-denoisers or latent planning to improve long-range coherence.\n\n- Architectural and planning aids:\n  - Retrieval-augmented generation to ground outputs.\n  - Tool use and structured prompting to constrain drift.\n  - Memory and planning modules, hierarchical decoding, or sentence-level control.\n\n- Prompting tips:\n  - Ask for concise answers, set token limits, and specify structure.\n  - Provide concrete constraints or content to reduce generic filler.\n  - Use “say nothing if uncertain” style instructions to avoid vacuity.\n\nRepresentative papers/terms to search:\n- Holtzman et al., “The Curious Case of Neural Text Degeneration” (2020): nucleus sampling.\n- Welleck et al., “Neural Text Degeneration with Unlikelihood Training.”\n- Li et al., “A Contrastive Framework for Decoding.”\n- Su et al., “DoLa: Decoding by Contrasting Layers.”\n- Meister et al., “Typical Decoding.”\n- Ouyang et al., “Training language models to follow instructions with human feedback.”\n\nIn short, degeneration arises from a mismatch between next-token likelihood and human preferences plus decoding choices; careful decoding, training objectives, and grounding help prevent it.",
          "type": "output_text",
          "logprobs": null
        }
      ],
      "role": "assistant",
      "status": "completed",
      "type": "message"
    }
  ],
  "parallel_tool_calls": true,
  "temperature": 1.0,
  "tool_choice": "auto",
  "tools": [],
  "top_p": 1.0,
  "background": false,
  "max_output_tokens": null,
  "max_tool_calls": null,
  "previous_response_id": null,
  "prompt": null,
  "prompt_cache_key": null,
  "reasoning": {
    "effort": "minimal",
    "generate_summary": null,
    "summary": "detailed"
  },
  "safety_identifier": null,
  "service_tier": "default",
  "status": "completed",
  "text": {
    "format": {
      "type": "text"
    }
  },
  "top_logprobs": null,
  "truncation": "disabled",
  "usage": {
    "input_tokens": 16,
    "input_tokens_details": {
      "cached_tokens": 0
    },
    "output_tokens": 657,
    "output_tokens_details": {
      "reasoning_tokens": 0
    },
    "total_tokens": 673
  },
  "user": null,
  "content_filters": null,
  "store": true
}

Python lark

GPT-5 시리즈 추론 모델은 custom_tool이라는 새로운 lark_tool을 호출할 수 있는 기능을 가지고 있습니다. 이 도구는 Python lark를 기반으로 하며 모델 출력을 보다 유연하게 제한하는 데 사용할 수 있습니다.

응답 API

{
  "model": "gpt-5-2025-08-07",
  "input": "please calculate the area of a circle with radius equal to the number of 'r's in strawberry",
  "tools": [
    {
      "type": "custom",
      "name": "lark_tool",
      "format": {
        "type": "grammar",
        "syntax": "lark",
        "definition": "start: QUESTION NEWLINE ANSWER\nQUESTION: /[^\\n?]{1,200}\\?/\nNEWLINE: /\\n/\nANSWER: /[^\\n!]{1,200}!/"
      }
    }
  ],
  "tool_choice": "required"
}

import os
from openai import OpenAI

client = OpenAI(  
  base_url = "https://YOUR-RESOURCE-NAME.openai.azure.com/openai/v1/",
  api_key=os.getenv("AZURE_OPENAI_API_KEY")  
)

response = client.responses.create(  
    model="gpt-5",  # replace with your model deployment name  
    tools=[  
        {  
            "type": "custom",
            "name": "lark_tool",
            "format": {
                "type": "grammar",
                "syntax": "lark",
                "definition": "start: QUESTION NEWLINE ANSWER\nQUESTION: /[^\\n?]{1,200}\\?/\nNEWLINE: /\\n/\nANSWER: /[^\\n!]{1,200}!/"
            }
        }  
    ],  
    input=[{"role": "user", "content": "Please calculate the area of a circle with radius equal to the number of 'r's in strawberry"}],  
)  

print(response.model_dump_json(indent=2))  
  

출력:

{
  "id": "resp_689a0cf927408190b8875915747667ad01c936c6ffb9d0d3",
  "created_at": 1754926332.0,
  "error": null,
  "incomplete_details": null,
  "instructions": null,
  "metadata": {},
  "model": "gpt-5",
  "object": "response",
  "output": [
    {
      "id": "rs_689a0cfd1c888190a2a67057f471b5cc01c936c6ffb9d0d3",
      "summary": [],
      "type": "reasoning",
      "encrypted_content": null,
      "status": null
    },
    {
      "id": "msg_689a0d00e60c81908964e5e9b2d6eeb501c936c6ffb9d0d3",
      "content": [
        {
          "annotations": [],
          "text": "“strawberry” has 3 r’s, so the radius is 3.\nArea = πr² = π × 3² = 9π ≈ 28.27 square units.",
          "type": "output_text",
          "logprobs": null
        }
      ],
      "role": "assistant",
      "status": "completed",
      "type": "message"
    }
  ],
  "parallel_tool_calls": true,
  "temperature": 1.0,
  "tool_choice": "auto",
  "tools": [
    {
      "name": "lark_tool",
      "parameters": null,
      "strict": null,
      "type": "custom",
      "description": null,
      "format": {
        "type": "grammar",
        "definition": "start: QUESTION NEWLINE ANSWER\nQUESTION: /[^\\n?]{1,200}\\?/\nNEWLINE: /\\n/\nANSWER: /[^\\n!]{1,200}!/",
        "syntax": "lark"
      }
    }
  ],
  "top_p": 1.0,
  "background": false,
  "max_output_tokens": null,
  "max_tool_calls": null,
  "previous_response_id": null,
  "prompt": null,
  "prompt_cache_key": null,
  "reasoning": {
    "effort": "medium",
    "generate_summary": null,
    "summary": null
  },
  "safety_identifier": null,
  "service_tier": "default",
  "status": "completed",
  "text": {
    "format": {
      "type": "text"
    }
  },
  "top_logprobs": null,
  "truncation": "disabled",
  "usage": {
    "input_tokens": 139,
    "input_tokens_details": {
      "cached_tokens": 0
    },
    "output_tokens": 240,
    "output_tokens_details": {
      "reasoning_tokens": 192
    },
    "total_tokens": 379
  },
  "user": null,
  "content_filters": null,
  "store": true
}

채팅 완료

{
  "messages": [
    {
      "role": "user",
      "content": "Which one is larger, 42 or 0?"
    }
  ],
  "tools": [
    {
      "type": "custom",
      "name": "custom_tool",
      "custom": {
        "name": "lark_tool",
        "format": {
          "type": "grammar",
          "grammar": {
            "syntax": "lark",
            "definition": "start: QUESTION NEWLINE ANSWER\nQUESTION: /[^\\n?]{1,200}\\?/\nNEWLINE: /\\n/\nANSWER: /[^\\n!]{1,200}!/"
          }
        }
      }
    }
  ],
  "tool_choice": "required",
  "model": "gpt-5-2025-08-07"
}

Markdown 출력

기본적으로 o3-mini 모델과 o1 모델은 markdown 서식을 포함하는 출력을 생성하려고 시도하지 않습니다. 이 동작이 바람직하지 않은 일반적인 사용 사례는 모델이 markdown 코드 블록 내에 포함된 코드를 출력하려는 경우입니다. 모델이 markdown 서식 지정 없이 출력을 생성하면 구문 강조 표시와 같은 기능과 대화형 플레이그라운드 환경에서 복사 가능한 코드 블록이 손실됩니다. 이 새로운 기본 동작을 재정의하고 모델 응답에 markdown을 포함하도록 하려면 개발자 메시지의 시작 부분에 문자열 Formatting re-enabled 을 추가합니다.

개발자 메시지의 시작 부분에 추가 Formatting re-enabled 해도 모델에 응답에 markdown 서식이 포함된다는 보장은 없으며 가능성만 높아집니다. 우리는 내부 테스트에서 Formatting re-enabled 모델 자체 o1 에 비해 o3-mini덜 효과적인 것을 발견했습니다.

성능을 Formatting re-enabled 향상시키려면 개발자 메시지의 시작을 더욱 확대하여 원하는 출력을 생성할 수 있습니다. 개발자 메시지의 시작 부분에 추가하는 Formatting re-enabled 대신 아래 예제 중 하나와 같이 보다 설명적인 초기 명령을 추가하는 방법을 실험해 볼 수 있습니다.

• Formatting re-enabled - please enclose code blocks with appropriate markdown tags.
• Formatting re-enabled - code output should be wrapped in markdown.

예상 출력에 따라 특정 사용 사례를 대상으로 하려면 초기 개발자 메시지를 추가로 사용자 지정해야 할 수 있습니다.