Strukturerade utdata

2025-10-10

Strukturerade utdata gör att en modell följer en JSON-schemadefinition som du anger som en del av api-anropet för slutsatsdragning. Detta står i kontrast till den äldre JSON-lägesfunktionen , som garanterade att giltig JSON skulle genereras, men inte kunde säkerställa strikt efterlevnad av det angivna schemat. Strukturerade utdata rekommenderas för funktionsanrop, extrahering av strukturerade data och skapande av komplexa arbetsflöden i flera steg.

Anmärkning

För närvarande stöds inte strukturerade utdata med:

Ta med dina egna datascenarier .
Assistenter eller Azure AI Agents Service.
gpt-4o-audio-preview och gpt-4o-mini-audio-preview version: 2024-12-17.

Modeller som stöds

gpt-5-pro version 2025-10-06
gpt-5-codex version 2025-09-11
gpt-5 version 2025-08-07
gpt-5-mini version 2025-08-07
gpt-5-nano version 2025-08-07
codex-mini version 2025-05-16
o3-pro version 2025-06-10
o3-mini version 2025-01-31
o1 version: 2024-12-17
gpt-4o-mini version: 2024-07-18
gpt-4o version: 2024-08-06
gpt-4o version: 2024-11-20
gpt-4.1 version 2025-04-14
gpt-4.1-nano version 2025-04-14
gpt-4.1-mini version: 2025-04-14
o4-mini version: 2025-04-16
o3 version: 2025-04-16

API-stöd

Stöd för strukturerade utdata lades först till i API-versionen 2024-08-01-preview. Den är tillgänglig i de senaste förhandsversions-API:erna samt det senaste GA-API:et: v1.

Du kan använda Pydantic för att definiera objektscheman i Python. Beroende på vilken version av OpenAI och Pydantic bibliotek du kör kan du behöva uppgradera till en nyare version. De här exemplen testades mot openai 1.42.0 och pydantic 2.8.2.

pip install openai pydantic --upgrade

Om du inte har använt Microsoft Entra-ID för autentisering tidigare kan du läsa Konfigurera Azure OpenAI i Azure AI Foundry Models med Microsoft Entra ID-autentisering.

import os
from pydantic import BaseModel
from openai import OpenAI
from azure.identity import DefaultAzureCredential, get_bearer_token_provider

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

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

class CalendarEvent(BaseModel):
    name: str
    date: str
    participants: list[str]

completion = client.beta.chat.completions.parse(
    model="MODEL_DEPLOYMENT_NAME", # replace with the model deployment name of your gpt-4o 2024-08-06 deployment
    messages=[
        {"role": "system", "content": "Extract the event information."},
        {"role": "user", "content": "Alice and Bob are going to a science fair on Friday."},
    ],
    response_format=CalendarEvent,
)

event = completion.choices[0].message.parsed

print(event)
print(completion.model_dump_json(indent=2))

Utgång

name='Science Fair' date='Friday' participants=['Alice', 'Bob']
{
  "id": "chatcmpl-A1EUP2fAmL4SeB1lVMinwM7I2vcqG",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "logprobs": null,
      "message": {
        "content": "{\n  \"name\": \"Science Fair\",\n  \"date\": \"Friday\",\n  \"participants\": [\"Alice\", \"Bob\"]\n}",
        "refusal": null,
        "role": "assistant",
        "function_call": null,
        "tool_calls": [],
        "parsed": {
          "name": "Science Fair",
          "date": "Friday",
          "participants": [
            "Alice",
            "Bob"
          ]
        }
      }
    }
  ],
  "created": 1724857389,
  "model": "gpt-4o-2024-08-06",
  "object": "chat.completion",
  "service_tier": null,
  "system_fingerprint": "fp_1c2eaec9fe",
  "usage": {
    "completion_tokens": 27,
    "prompt_tokens": 32,
    "total_tokens": 59
  }
}

pip install openai pydantic --upgrade

import os
from pydantic import BaseModel
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")  
)

class CalendarEvent(BaseModel):
    name: str
    date: str
    participants: list[str]

completion = client.beta.chat.completions.parse(
    model="MODEL_DEPLOYMENT_NAME", # replace with the model deployment name of your gpt-4o 2024-08-06 deployment
    messages=[
        {"role": "system", "content": "Extract the event information."},
        {"role": "user", "content": "Alice and Bob are going to a science fair on Friday."},
    ],
    response_format=CalendarEvent,
)

event = completion.choices[0].message.parsed

print(event)
print(completion.model_dump_json(indent=2))

Utgång

name='Science Fair' date='Friday' participants=['Alice', 'Bob']
{
  "id": "chatcmpl-A1EUP2fAmL4SeB1lVMinwM7I2vcqG",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "logprobs": null,
      "message": {
        "content": "{\n  \"name\": \"Science Fair\",\n  \"date\": \"Friday\",\n  \"participants\": [\"Alice\", \"Bob\"]\n}",
        "refusal": null,
        "role": "assistant",
        "function_call": null,
        "tool_calls": [],
        "parsed": {
          "name": "Science Fair",
          "date": "Friday",
          "participants": [
            "Alice",
            "Bob"
          ]
        }
      }
    }
  ],
  "created": 1724857389,
  "model": "gpt-4o-2024-08-06",
  "object": "chat.completion",
  "service_tier": null,
  "system_fingerprint": "fp_1c2eaec9fe",
  "usage": {
    "completion_tokens": 27,
    "prompt_tokens": 32,
    "total_tokens": 59
  }
}

Funktionsanrop med strukturerade utdata

Strukturerade utdata kan aktiveras för funktionsanrop med en enda parameter, genom att ange strict: true.

Anmärkning

Strukturerade utdata stöds inte med parallella funktionsanrop. När du använder strukturerade utdata, ställ in parallel_tool_calls till false.

Python (Microsoft Entra-ID)
Python (nyckelbaserad autentisering)

from enum import Enum
from typing import Union
from pydantic import BaseModel
from openai import OpenAI
from azure.identity import DefaultAzureCredential, get_bearer_token_provider

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

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

class GetDeliveryDate(BaseModel):
    order_id: str

tools = [openai.pydantic_function_tool(GetDeliveryDate)]

messages = []
messages.append({"role": "system", "content": "You are a helpful customer support assistant. Use the supplied tools to assist the user."})
messages.append({"role": "user", "content": "Hi, can you tell me the delivery date for my order #12345?"}) 

response = client.chat.completions.create(
    model="MODEL_DEPLOYMENT_NAME", # replace with the model deployment name of your gpt-4o 2024-08-06 deployment
    messages=messages,
    tools=tools
)

print(response.choices[0].message.tool_calls[0].function)
print(response.model_dump_json(indent=2))

from enum import Enum
from typing import Union
from pydantic import BaseModel
import openai
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")  
)


class GetDeliveryDate(BaseModel):
    order_id: str

tools = [openai.pydantic_function_tool(GetDeliveryDate)]

messages = []
messages.append({"role": "system", "content": "You are a helpful customer support assistant. Use the supplied tools to assist the user."})
messages.append({"role": "user", "content": "Hi, can you tell me the delivery date for my order #12345?"}) 

response = client.chat.completions.create(
    model="MODEL_DEPLOYMENT_NAME", # replace with the model deployment name of your gpt-4o 2024-08-06 deployment
    messages=messages,
    tools=tools
)

print(response.choices[0].message.tool_calls[0].function)
print(response.model_dump_json(indent=2))

Komma igång

Lägg till följande paket i projektet för att arbeta med Azure OpenAI:

Azure.AI.OpenAI: Tillhandahåller en Azure OpenAI-klient med Azure-specifika funktioner som bygger på standardberoendet för OpenAI-bibliotek .
Azure.Identity: Tillhandahåller stöd för autentisering av Microsoft Entra-ID-token i Azure SDK-biblioteken.
Newtonsoft.Json.Schema: Tillhandahåller användbara verktyg för att arbeta med JSON-scheman.

dotnet add package Azure.AI.OpenAI --prerelease
dotnet add package Azure.Identity
dotnet add package Newtonsoft.Json.Schema

Microsoft Entra-ID
Nyckelbaserad autentisering

Om du inte har använt Microsoft Entra-ID för autentisering tidigare kan du läsa Konfigurera Azure OpenAI i Azure AI Foundry Models med Microsoft Entra ID-autentisering.

using Azure.AI.OpenAI;
using Azure.Identity;
using Newtonsoft.Json.Schema.Generation;
using OpenAI.Chat;
using System.ClientModel;

// Create the clients
string endpoint = GetEnvironmentVariable("AZURE_OPENAI_ENDPOINT");

AzureOpenAIClient openAIClient = new(
    new Uri(endpoint),
    new DefaultAzureCredential());

var client = openAIClient.GetChatClient("gpt-4o");

// Create a chat with initial prompts
var chat = new List<ChatMessage>()
    {
         new SystemChatMessage("Extract the event information and projected weather."),
         new UserChatMessage("Alice and Bob are going to a science fair in Seattle on June 1st, 2025.")
    };

// Get the schema of the class for the structured response
JSchemaGenerator generator = new JSchemaGenerator();
var jsonSchema = generator.Generate(typeof(CalendarEvent)).ToString();

// Get a completion with structured output
var chatUpdates = client.CompleteChatStreamingAsync(
        chat,
        new ChatCompletionOptions()
        {
            ResponseFormat = ChatResponseFormat.CreateJsonSchemaFormat(
                        "calenderEvent",
                        BinaryData.FromString(jsonSchema))
        });

// Write the structured response
await foreach (var chatUpdate in chatUpdates)
{
    foreach (var contentPart in chatUpdate.ContentUpdate)
    {
        Console.Write(contentPart.Text);
    }
}

// The class for the structured response
public class CalendarEvent()
{
    public string Name { get; set; }
    public string Date { get; set; }
    public List<string> Participants { get; set; }
}

using Azure.AI.OpenAI;
using Newtonsoft.Json.Schema.Generation;
using OpenAI.Chat;
using System.ClientModel;

// Create the clients
string endpoint = GetEnvironmentVariable("AZURE_OPENAI_ENDPOINT");
string key = GetEnvironmentVariable("AZURE_OPENAI_API_KEY");

AzureOpenAIClient openAIClient = new(
    new Uri(endpoint),
    new ApiKeyCredential(key));

var client = openAIClient.GetChatClient("gpt-4o");

// Create a chat with initial prompts
var chat = new List<ChatMessage>()
    {
         new SystemChatMessage("Extract the event information and projected weather."),
         new UserChatMessage("Alice and Bob are going to a science fair in Seattle on June 1st, 2025.")
    };

// Get the schema of the class for the structured response
JSchemaGenerator generator = new JSchemaGenerator();
var jsonSchema = generator.Generate(typeof(CalendarEvent)).ToString();

// Get a completion with structured output
var chatUpdates = client.CompleteChatStreamingAsync(
        chat,
        new ChatCompletionOptions()
        {
            ResponseFormat = ChatResponseFormat.CreateJsonSchemaFormat(
                        "calenderEvent",
                        BinaryData.FromString(jsonSchema))
        });

// Write the structured response
await foreach (var chatUpdate in chatUpdates)
{
    foreach (var contentPart in chatUpdate.ContentUpdate)
    {
        Console.Write(contentPart.Text);
    }
}

// The class for the structured response
public class CalendarEvent()
{
    public string Name { get; set; }
    public string Date { get; set; }
    public List<string> Participants { get; set; }
}

Funktionsanrop med strukturerade utdata

Strukturerade utdata kan aktiveras för funktionsanrop med en enda parameter, genom att ange strict: true.

Microsoft Entra-ID
Nyckelbaserad autentisering

using Azure.AI.OpenAI;
using Newtonsoft.Json.Schema.Generation;
using OpenAI.Chat;
using System.ClientModel;

// Create the clients
string endpoint = GetEnvironmentVariable("AZURE_OPENAI_ENDPOINT");

AzureOpenAIClient openAIClient = new(
    new Uri(endpoint),
    new DefaultAzureCredential());

var chatClient = openAIClient.GetChatClient("gpt-4o");

// Local function to be used by the assistant tooling
string GetTemperature(string location, string date)
{
    // Placeholder for Weather API
    if(location == "Seattle" && date == "2025-06-01")
    {
        return "75";
    }

    return "50";
}

// Create a tool to get the temperature
ChatTool GetTemperatureTool = ChatTool.CreateFunctionTool(
    functionName: nameof(GetTemperature),
    functionSchemaIsStrict: true,
    functionDescription: "Get the projected temperature by date and location.",
    functionParameters: BinaryData.FromBytes("""
        {
            "type": "object",
            "properties": {
                "location": {
                    "type": "string",
                    "description": "The location of the weather."
                },
                "date": {
                    "type": "string",
                    "description": "The date of the projected weather."
                }
            },
            "required": ["location", "date"],
            "additionalProperties": false  
        }
        """u8.ToArray())
);

// Create a chat with prompts
var chat = new List<ChatMessage>()
    {
         new SystemChatMessage("Extract the event information and projected weather."),
         new UserChatMessage("Alice and Bob are going to a science fair in Seattle on June 1st, 2025.")
    };

// Create a JSON schema for the CalendarEvent structured response
JSchemaGenerator generator = new JSchemaGenerator();
var jsonSchema = generator.Generate(typeof(CalendarEvent)).ToString();

// Get a chat completion from the AI model
var completion = chatClient.CompleteChat(
        chat,
        new ChatCompletionOptions()
        {
            ResponseFormat = ChatResponseFormat.CreateJsonSchemaFormat(
                "calenderEvent",
                BinaryData.FromString(jsonSchema)),
            Tools = { GetTemperatureTool }
        });

Console.WriteLine(completion.Value.ToolCalls[0].FunctionName);

// Structured response class
public class CalendarEvent()
{
    public string Name { get; set; }
    public string Date { get; set; }
    public string Temperature { get; set; }
    public List<string> Participants { get; set; }
}

using Azure.AI.OpenAI;
using Newtonsoft.Json.Schema.Generation;
using OpenAI.Chat;
using System.ClientModel;

// Create the clients
string endpoint = GetEnvironmentVariable("AZURE_OPENAI_ENDPOINT");
string key = GetEnvironmentVariable("AZURE_OPENAI_API_KEY");

AzureOpenAIClient openAIClient = new(
    new Uri(endpoint),
    new ApiKeyCredential(key));

var chatClient = openAIClient.GetChatClient("gpt-4o");

// Local function to be used by the assistant tooling
string GetTemperature(string location, string date)
{
    // Placeholder for Weather API
    if(location == "Seattle" && date == "2025-06-01")
    {
        return "75";
    }

    return "50";
}

// Create a tool to get the temperature
ChatTool GetTemperatureTool = ChatTool.CreateFunctionTool(
    functionName: nameof(GetTemperature),
    functionSchemaIsStrict: true,
    functionDescription: "Get the projected temperature by date and location.",
    functionParameters: BinaryData.FromBytes("""
        {
            "type": "object",
            "properties": {
                "location": {
                    "type": "string",
                    "description": "The location of the weather."
                },
                "date": {
                    "type": "string",
                    "description": "The date of the projected weather."
                }
            },
            "required": ["location", "date"],
            "additionalProperties": false  
        }
        """u8.ToArray())
);

// Create a chat with prompts
var chat = new List<ChatMessage>()
    {
         new SystemChatMessage("Extract the event information and projected weather."),
         new UserChatMessage("Alice and Bob are going to a science fair in Seattle on June 1st, 2025.")
    };

// Create a JSON schema for the CalendarEvent structured response
JSchemaGenerator generator = new JSchemaGenerator();
var jsonSchema = generator.Generate(typeof(CalendarEvent)).ToString();

// Get a chat completion from the AI model
var completion = chatClient.CompleteChat(
        chat,
        new ChatCompletionOptions()
        {
            ResponseFormat = ChatResponseFormat.CreateJsonSchemaFormat(
                "calenderEvent",
                BinaryData.FromString(jsonSchema)),
            Tools = { GetTemperatureTool }
        });

Console.WriteLine(completion.Value.ToolCalls[0].FunctionName);

// Structured response class
public class CalendarEvent()
{
    public string Name { get; set; }
    public string Date { get; set; }
    public string Temperature { get; set; }
    public List<string> Participants { get; set; }
}

Komma igång

response_format är inställt på json_schema med strict: true inställt.

curl -X POST  https://YOUR_RESOURCE_NAME.openai.azure.com/openai/v1/chat/completions \
  -H "api-key: $AZURE_OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
    -d '{
        "model": "YOUR_MODEL_DEPLOYMENT_NAME",
        "messages": [
                {"role": "system", "content": "Extract the event information."},
                {"role": "user", "content": "Alice and Bob are going to a science fair on Friday."}
            ],
            "response_format": {
                "type": "json_schema",
                "json_schema": {
                    "name": "CalendarEventResponse",
                    "strict": true,
                    "schema": {
                        "type": "object",
                        "properties": {
                            "name": {
                              "type": "string"
                            },
                            "date": {
                                "type": "string"
                            },
                            "participants": {
                                "type": "array",
                                "items": {
                                    "type": "string"
                                }
                            }
                        },
                        "required": [
                            "name",
                            "date",
                            "participants"
                        ],
                        "additionalProperties": false
                    }
                }
          }
  }'

Utdata:

{
  "id": "chatcmpl-A1HKsHAe2hH9MEooYslRn9UmEwsag",
  "object": "chat.completion",
  "created": 1724868330,
  "model": "gpt-4o-2024-08-06",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "{\n  \"name\": \"Science Fair\",\n  \"date\": \"Friday\",\n  \"participants\": [\"Alice\", \"Bob\"]\n}"
      },
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 33,
    "completion_tokens": 27,
    "total_tokens": 60
  },
  "system_fingerprint": "fp_1c2eaec9fe"
}

Funktionsanrop med strukturerade utdata

curl -X POST  https://YOUR_RESOURCE_NAME.openai.azure.com/openai/v1/chat/completions \
  -H "api-key: $AZURE_OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "YOUR_MODEL_DEPLOYMENT_NAME",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant. The current date is August 6, 2024. You help users query for the data they are looking for by calling the query function."
    },
    {
      "role": "user",
      "content": "look up all my orders in may of last year that were fulfilled but not delivered on time"
    }
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "query",
        "description": "Execute a query.",
        "strict": true,
        "parameters": {
          "type": "object",
          "properties": {
            "table_name": {
              "type": "string",
              "enum": ["orders"]
            },
            "columns": {
              "type": "array",
              "items": {
                "type": "string",
                "enum": [
                  "id",
                  "status",
                  "expected_delivery_date",
                  "delivered_at",
                  "shipped_at",
                  "ordered_at",
                  "canceled_at"
                ]
              }
            },
            "conditions": {
              "type": "array",
              "items": {
                "type": "object",
                "properties": {
                  "column": {
                    "type": "string"
                  },
                  "operator": {
                    "type": "string",
                    "enum": ["=", ">", "<", ">=", "<=", "!="]
                  },
                  "value": {
                    "anyOf": [
                      {
                        "type": "string"
                      },
                      {
                        "type": "number"
                      },
                      {
                        "type": "object",
                        "properties": {
                          "column_name": {
                            "type": "string"
                          }
                        },
                        "required": ["column_name"],
                        "additionalProperties": false
                      }
                    ]
                  }
                },
                "required": ["column", "operator", "value"],
                "additionalProperties": false
              }
            },
            "order_by": {
              "type": "string",
              "enum": ["asc", "desc"]
            }
          },
          "required": ["table_name", "columns", "conditions", "order_by"],
          "additionalProperties": false
        }
      }
    }
  ]
}'

Scheman och begränsningar som stöds

Azure OpenAI-strukturerade utdata stöder samma delmängd av JSON-schemat som OpenAI.

Typer som stöds

Sträng
Nummer
Boolean
Heltal
Objekt
Array
Enum
någon av

Anmärkning

Rotobjekt kan inte vara typen anyOf .

Alla fält måste vara obligatoriska

Alla fält eller funktionsparametrar måste inkluderas efter behov. I exemplet nedan location, och unit anges båda under "required": ["location", "unit"].

{
    "name": "get_weather",
    "description": "Fetches the weather in the given location",
    "strict": true,
    "parameters": {
        "type": "object",
        "properties": {
            "location": {
                "type": "string",
                "description": "The location to get the weather for"
            },
            "unit": {
                "type": "string",
                "description": "The unit to return the temperature in",
                "enum": ["F", "C"]
            }
        },
        "additionalProperties": false,
        "required": ["location", "unit"]
    }

Om det behövs kan du emulera en valfri parameter med hjälp av en union-typ med null. I det här exemplet uppnås detta med raden "type": ["string", "null"],.

{
    "name": "get_weather",
    "description": "Fetches the weather in the given location",
    "strict": true,
    "parameters": {
        "type": "object",
        "properties": {
            "location": {
                "type": "string",
                "description": "The location to get the weather for"
            },
            "unit": {
                "type": ["string", "null"],
                "description": "The unit to return the temperature in",
                "enum": ["F", "C"]
            }
        },
        "additionalProperties": false,
        "required": [
            "location", "unit"
        ]
    }
}

Nästningsdjup

Ett schema kan ha upp till 100 objektegenskaper totalt, med upp till fem kapslingsnivåer

additionalProperties: false måste alltid anges i objekt

Den här egenskapen styr om ett objekt kan ha ytterligare nyckelvärdepar som inte har definierats i JSON-schemat. För att kunna använda strukturerade utdata måste du ange värdet falskt.

Nyckelordning

Strukturerade utdata sorteras på samma sätt som det angivna schemat. Om du vill ändra utdataordningen ändrar du ordningen på schemat som du skickar som en del av din slutsatsdragningsbegäran.

Typspecifika nyckelord som inte stöds

Typ	Nyckelord som inte stöds
Sträng	minlängd maxlängd mönster format
Nummer	minsta maximal multipleOf
Objekt	mönsterEgenskaper outvärderadeEgenskaper egenskapsnamn minEgenskaper maximalt antal egenskaper
Matriser	oevaluerade objekt Innehåller minInnehåller maxInnehåller minItems maxItems unikaFöremål

Kapslade scheman med anyOf måste följa den övergripande JSON-schemadeluppsättningen

Exempel på schema som stöds anyOf :

{
    "type": "object",
    "properties": {
        "item": {
            "anyOf": [
                {
                    "type": "object",
                    "description": "The user object to insert into the database",
                    "properties": {
                        "name": {
                            "type": "string",
                            "description": "The name of the user"
                        },
                        "age": {
                            "type": "number",
                            "description": "The age of the user"
                        }
                    },
                    "additionalProperties": false,
                    "required": [
                        "name",
                        "age"
                    ]
                },
                {
                    "type": "object",
                    "description": "The address object to insert into the database",
                    "properties": {
                        "number": {
                            "type": "string",
                            "description": "The number of the address. Eg. for 123 main st, this would be 123"
                        },
                        "street": {
                            "type": "string",
                            "description": "The street name. Eg. for 123 main st, this would be main st"
                        },
                        "city": {
                            "type": "string",
                            "description": "The city of the address"
                        }
                    },
                    "additionalProperties": false,
                    "required": [
                        "number",
                        "street",
                        "city"
                    ]
                }
            ]
        }
    },
    "additionalProperties": false,
    "required": [
        "item"
    ]
}

Definitioner stöds

Exempel som stöds:

{
    "type": "object",
    "properties": {
        "steps": {
            "type": "array",
            "items": {
                "$ref": "#/$defs/step"
            }
        },
        "final_answer": {
            "type": "string"
        }
    },
    "$defs": {
        "step": {
            "type": "object",
            "properties": {
                "explanation": {
                    "type": "string"
                },
                "output": {
                    "type": "string"
                }
            },
            "required": [
                "explanation",
                "output"
            ],
            "additionalProperties": false
        }
    },
    "required": [
        "steps",
        "final_answer"
    ],
    "additionalProperties": false
}

Rekursiva scheman stöds

Exempel med # för rotrekursion:

{
        "name": "ui",
        "description": "Dynamically generated UI",
        "strict": true,
        "schema": {
            "type": "object",
            "properties": {
                "type": {
                    "type": "string",
                    "description": "The type of the UI component",
                    "enum": ["div", "button", "header", "section", "field", "form"]
                },
                "label": {
                    "type": "string",
                    "description": "The label of the UI component, used for buttons or form fields"
                },
                "children": {
                    "type": "array",
                    "description": "Nested UI components",
                    "items": {
                        "$ref": "#"
                    }
                },
                "attributes": {
                    "type": "array",
                    "description": "Arbitrary attributes for the UI component, suitable for any element",
                    "items": {
                        "type": "object",
                        "properties": {
                            "name": {
                                "type": "string",
                                "description": "The name of the attribute, for example onClick or className"
                            },
                            "value": {
                                "type": "string",
                                "description": "The value of the attribute"
                            }
                        },
                      "additionalProperties": false,
                      "required": ["name", "value"]
                    }
                }
            },
            "required": ["type", "label", "children", "attributes"],
            "additionalProperties": false
        }
    }

Exempel på explicit rekursion:

{
    "type": "object",
    "properties": {
        "linked_list": {
            "$ref": "#/$defs/linked_list_node"
        }
    },
    "$defs": {
        "linked_list_node": {
            "type": "object",
            "properties": {
                "value": {
                    "type": "number"
                },
                "next": {
                    "anyOf": [
                        {
                            "$ref": "#/$defs/linked_list_node"
                        },
                        {
                            "type": "null"
                        }
                    ]
                }
            },
            "additionalProperties": false,
            "required": [
                "next",
                "value"
            ]
        }
    },
    "additionalProperties": false,
    "required": [
        "linked_list"
    ]
}

Feedback

Var den här sidan till hjälp?

Dela via

Strukturerade utdata

Modeller som stöds

API-stöd

Komma igång

Utgång

Funktionsanrop med strukturerade utdata

Komma igång

Funktionsanrop med strukturerade utdata

Komma igång

Funktionsanrop med strukturerade utdata

Scheman och begränsningar som stöds

Typer som stöds

Alla fält måste vara obligatoriska

Nästningsdjup

additionalProperties: false måste alltid anges i objekt

Nyckelordning

Typspecifika nyckelord som inte stöds

Kapslade scheman med anyOf måste följa den övergripande JSON-schemadeluppsättningen

Definitioner stöds

Rekursiva scheman stöds

Feedback

Ytterligare resurser