Skip to main content

结构化输出 (JSON 模式)

快速开始​

from litellm import completion
import os 

os.environ["OPENAI_API_KEY"] = ""

response = completion(
  model="gpt-4o-mini",
  response_format={ "type": "json_object" },
  messages=[
    {"role": "system", "content": "You are a helpful assistant designed to output JSON."},
    {"role": "user", "content": "Who won the world series in 2020?"}
  ]
)
print(response.choices[0].message.content)
curl http://0.0.0.0:4000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $LITELLM_KEY" \
  -d '{
    "model": "gpt-4o-mini",
    "response_format": { "type": "json_object" },
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful assistant designed to output JSON."
      },
      {
        "role": "user",
        "content": "Who won the world series in 2020?"
      }
    ]
  }'

检查模型支持​

调用 litellm.get_supported_openai_params 来检查某个模型/提供者是否支持 response_format。

from litellm import get_supported_openai_params

params = get_supported_openai_params(model="anthropic.claude-3", custom_llm_provider="bedrock")

assert "response_format" in params

传递 'json_schema'​

要使用结构化输出，只需指定

response_format: { "type": "json_schema", "json_schema": … , "strict": true }

适用于：

  • OpenAI 模型
  • Azure OpenAI 模型
  • Google AI Studio - Gemini 模型
  • Vertex AI 模型 (Gemini + Anthropic)
  • Bedrock 模型
import os
from litellm import completion 
from pydantic import BaseModel

# 添加到环境变量
os.environ["OPENAI_API_KEY"] = ""

messages = [{"role": "user", "content": "List 5 important events in the XIX century"}]

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

class EventsList(BaseModel):
    events: list[CalendarEvent]

resp = completion(
    model="gpt-4o-2024-08-06",
    messages=messages,
    response_format=EventsList
)

print("Received={}".format(resp))
  1. 将 openai 模型添加到 config.yaml
model_list:
  - model_name: "gpt-4o"
    litellm_params:
      model: "gpt-4o-2024-08-06"
  1. 使用 config.yaml 启动代理
litellm --config /path/to/config.yaml
  1. 使用 OpenAI SDK / Curl 调用！

只需替换 openai sdk 中的 'base_url'，即可使用 'json_schema' 调用 openai 模型的代理

OpenAI SDK

from pydantic import BaseModel
from openai import OpenAI

client = OpenAI(
    api_key="anything", # 👈 PROXY KEY (可以是任何值，如果未设置 master_key)
    base_url="http://0.0.0.0:4000" # 👈 PROXY BASE URL
)

class Step(BaseModel):
    explanation: str
    output: str

class MathReasoning(BaseModel):
    steps: list[Step]
    final_answer: str

completion = client.beta.chat.completions.parse(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "You are a helpful math tutor. Guide the user through the solution step by step."},
        {"role": "user", "content": "how can I solve 8x + 7 = -23"}
    ],
    response_format=MathReasoning,
)

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

Curl

curl -X POST 'http://0.0.0.0:4000/v1/chat/completions' \
-H 'Content-Type: application/json' \
-H 'Authorization: Bearer sk-1234' \
-d '{
    "model": "gpt-4o",
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful math tutor. Guide the user through the solution step by step."
      },
      {
        "role": "user",
        "content": "how can I solve 8x + 7 = -23"
      }
    ],
    "response_format": {
      "type": "json_schema",
      "json_schema": {
        "name": "math_reasoning",
        "schema": {
          "type": "object",
          "properties": {
            "steps": {
              "type": "array",
              "items": {
                "type": "object",
                "properties": {
                  "explanation": { "type": "string" },
                  "output": { "type": "string" }
                },
                "required": ["explanation", "output"],
                "additionalProperties": false
              }
            },
            "final_answer": { "type": "string" }
          },
          "required": ["steps", "final_answer"],
          "additionalProperties": false
        },
        "strict": true
      }
    }
  }'

验证 JSON Schema​

并非所有 vertex 模型都支持将 json_schema 传递给他们（例如 gemini-1.5-flash）。为了解决这个问题，LiteLLM 支持客户端对 json schema 进行验证。

litellm.enable_json_schema_validation=True

如果设置了 litellm.enable_json_schema_validation=True，LiteLLM 将使用 jsonvalidator 验证 JSON 响应。

查看代码

# !gcloud auth application-default login - 运行此命令以将 vertex 凭据添加到您的环境
import litellm, os
from litellm import completion 
from pydantic import BaseModel 


messages=[
        {"role": "system", "content": "提取事件信息。"},
        {"role": "user", "content": "Alice 和 Bob 将在周五去科学展览会。"},
    ]

litellm.enable_json_schema_validation = True
litellm.set_verbose = True # 查看 litellm 发出的原始请求

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

resp = completion(
    model="gemini/gemini-1.5-pro",
    messages=messages,
    response_format=CalendarEvent,
)

print("Received={}".format(resp))
  1. 创建 config.yaml
model_list:
  - model_name: "gemini-1.5-flash"
    litellm_params:
      model: "gemini/gemini-1.5-flash"
      api_key: os.environ/GEMINI_API_KEY

litellm_settings:
  enable_json_schema_validation: True
  1. 启动代理
litellm --config /path/to/config.yaml
  1. 测试它！
curl http://0.0.0.0:4000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $LITELLM_API_KEY" \
  -d '{
    "model": "gemini-1.5-flash",
    "messages": [
        {"role": "system", "content": "提取事件信息。"},
        {"role": "user", "content": "Alice 和 Bob 将在周五去科学展览会。"},
    ],
    "response_format": { 
        "type": "json_object",
        "response_schema": { 
            "type": "json_schema",
            "json_schema": {
              "name": "math_reasoning",
              "schema": {
                "type": "object",
                "properties": {
                  "steps": {
                    "type": "array",
                    "items": {
                      "type": "object",
                      "properties": {
                        "explanation": { "type": "string" },
                        "output": { "type": "string" }
                      },
                      "required": ["explanation", "output"],
                      "additionalProperties": false
                    }
                  },
                  "final_answer": { "type": "string" }
                },
                "required": ["steps", "final_answer"],
                "additionalProperties": false
              },
              "strict": true
            },
        }
    },
  }'
优云智算