结构化输出

vLLM 支持使用 xgrammar 或 guidance 作为后端生成结构化输出。本文档将向您展示一些用于生成结构化输出的不同选项的示例。

在线服务 (OpenAI API)¶

您可以使用 OpenAI 的 Completions 和 Chat API 生成结构化输出。

支持以下参数，必须作为额外参数添加

guided_choice: 输出将是这些选项中的一个。
guided_regex: 输出将遵循正则表达式模式。
guided_json: 输出将遵循 JSON schema。
guided_grammar: 输出将遵循上下文无关文法。
structural_tag: 在生成的文本中，遵循指定标签集内的 JSON schema。

您可以在兼容 OpenAI 的服务器页面上查看支持的完整参数列表。

兼容 OpenAI 的服务器默认支持结构化输出。您可以通过将 --guided-decoding-backend 标志设置为 vllm serve 来指定要使用的后端。默认后端是 auto，它将尝试根据请求的详细信息选择适当的后端。您也可以选择特定的后端以及一些选项。完整的选项集可在 vllm serve --help 文本中找到。

现在让我们看看每种情况的一个示例，从最简单的 guided_choice 开始。

from openai import OpenAI
client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="-",
)

completion = client.chat.completions.create(
    model="Qwen/Qwen2.5-3B-Instruct",
    messages=[
        {"role": "user", "content": "Classify this sentiment: vLLM is wonderful!"}
    ],
    extra_body={"guided_choice": ["positive", "negative"]},
)
print(completion.choices[0].message.content)

下一个示例展示了如何使用 guided_regex。其思想是根据一个简单的正则表达式模板生成一个电子邮件地址。

completion = client.chat.completions.create(
    model="Qwen/Qwen2.5-3B-Instruct",
    messages=[
        {
            "role": "user",
            "content": "Generate an example email address for Alan Turing, who works in Enigma. End in .com and new line. Example result: [email protected]\n",
        }
    ],
    extra_body={"guided_regex": r"\w+@\w+\.com\n", "stop": ["\n"]},
)
print(completion.choices[0].message.content)

结构化文本生成中最相关的特性之一是生成具有预定义字段和格式的有效 JSON 的选项。为此，我们可以通过两种不同的方式使用 guided_json 参数：

直接使用 JSON Schema
定义一个 Pydantic 模型，然后从中提取 JSON Schema（这通常是更简单的选项）。

下一个示例展示了如何将 guided_json 参数与 Pydantic 模型一起使用

from pydantic import BaseModel
from enum import Enum

class CarType(str, Enum):
    sedan = "sedan"
    suv = "SUV"
    truck = "Truck"
    coupe = "Coupe"

class CarDescription(BaseModel):
    brand: str
    model: str
    car_type: CarType

json_schema = CarDescription.model_json_schema()

completion = client.chat.completions.create(
    model="Qwen/Qwen2.5-3B-Instruct",
    messages=[
        {
            "role": "user",
            "content": "Generate a JSON with the brand, model and car_type of the most iconic car from the 90's",
        }
    ],
    extra_body={"guided_json": json_schema},
)
print(completion.choices[0].message.content)

提示

虽然不是严格必需的，但通常最好在提示词中指示 JSON schema 以及字段应如何填充。这在大多数情况下可以显著提高结果。

最后是 guided_grammar 选项，这可能是最难使用的，但它确实非常强大。它允许我们定义完整的语言，如 SQL 查询。它通过使用上下文无关的 EBNF 文法来工作。例如，我们可以用它来定义特定格式的简化 SQL 查询

simplified_sql_grammar = """
    root ::= select_statement

    select_statement ::= "SELECT " column " from " table " where " condition

    column ::= "col_1 " | "col_2 "

    table ::= "table_1 " | "table_2 "

    condition ::= column "= " number

    number ::= "1 " | "2 "
"""

completion = client.chat.completions.create(
    model="Qwen/Qwen2.5-3B-Instruct",
    messages=[
        {
            "role": "user",
            "content": "Generate an SQL query to show the 'username' and 'email' from the 'users' table.",
        }
    ],
    extra_body={"guided_grammar": simplified_sql_grammar},
)
print(completion.choices[0].message.content)

完整示例： examples/online_serving/openai_chat_completion_structured_outputs.py

实验性自动解析 (OpenAI API)¶

本节介绍 OpenAI beta 包装器，它封装了 client.chat.completions.create() 方法，提供了与 Python 特定类型更丰富的集成。

在撰写本文时 (openai==1.54.4)，这是 OpenAI 客户端库中的一项“beta”功能。代码参考可以在这里找到。

对于以下示例，vLLM 是使用 vllm serve meta-llama/Llama-3.1-8B-Instruct 设置的

这是一个简单的示例，演示如何使用 Pydantic 模型获取结构化输出

from pydantic import BaseModel
from openai import OpenAI

class Info(BaseModel):
    name: str
    age: int

client = OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="dummy")
completion = client.beta.chat.completions.parse(
    model="meta-llama/Llama-3.1-8B-Instruct",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "My name is Cameron, I'm 28. What's my name and age?"},
    ],
    response_format=Info,
    extra_body=dict(guided_decoding_backend="outlines"),
)

message = completion.choices[0].message
print(message)
assert message.parsed
print("Name:", message.parsed.name)
print("Age:", message.parsed.age)

输出

ParsedChatCompletionMessage[Testing](content='{"name": "Cameron", "age": 28}', refusal=None, role='assistant', audio=None, function_call=None, tool_calls=[], parsed=Testing(name='Cameron', age=28))
Name: Cameron
Age: 28

这是一个更复杂的示例，使用嵌套的 Pydantic 模型来处理逐步数学解法

from typing import List
from pydantic import BaseModel
from openai import OpenAI

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

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

client = OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="dummy")
completion = client.beta.chat.completions.parse(
    model="meta-llama/Llama-3.1-8B-Instruct",
    messages=[
        {"role": "system", "content": "You are a helpful expert math tutor."},
        {"role": "user", "content": "Solve 8x + 31 = 2."},
    ],
    response_format=MathResponse,
    extra_body=dict(guided_decoding_backend="outlines"),
)

message = completion.choices[0].message
print(message)
assert message.parsed
for i, step in enumerate(message.parsed.steps):
    print(f"Step #{i}:", step)
print("Answer:", message.parsed.final_answer)

输出

ParsedChatCompletionMessage[MathResponse](content='{ "steps": [{ "explanation": "First, let\'s isolate the term with the variable \'x\'. To do this, we\'ll subtract 31 from both sides of the equation.", "output": "8x + 31 - 31 = 2 - 31"}, { "explanation": "By subtracting 31 from both sides, we simplify the equation to 8x = -29.", "output": "8x = -29"}, { "explanation": "Next, let\'s isolate \'x\' by dividing both sides of the equation by 8.", "output": "8x / 8 = -29 / 8"}], "final_answer": "x = -29/8" }', refusal=None, role='assistant', audio=None, function_call=None, tool_calls=[], parsed=MathResponse(steps=[Step(explanation="First, let's isolate the term with the variable 'x'. To do this, we'll subtract 31 from both sides of the equation.", output='8x + 31 - 31 = 2 - 31'), Step(explanation='By subtracting 31 from both sides, we simplify the equation to 8x = -29.', output='8x = -29'), Step(explanation="Next, let's isolate 'x' by dividing both sides of the equation by 8.", output='8x / 8 = -29 / 8')], final_answer='x = -29/8'))
Step #0: explanation="First, let's isolate the term with the variable 'x'. To do this, we'll subtract 31 from both sides of the equation." output='8x + 31 - 31 = 2 - 31'
Step #1: explanation='By subtracting 31 from both sides, we simplify the equation to 8x = -29.' output='8x = -29'
Step #2: explanation="Next, let's isolate 'x' by dividing both sides of the equation by 8." output='8x / 8 = -29 / 8'
Answer: x = -29/8

使用 structural_tag 的示例可以在此处找到： examples/online_serving/openai_chat_completion_structured_outputs_structural_tag.py

离线推理¶

离线推理支持相同类型的引导式解码。要使用它，我们需要在 SamplingParams 中使用 GuidedDecodingParams 类配置引导式解码。GuidedDecodingParams 中可用的主要选项有

json
regex
choice
grammar
structural_tag

这些参数的使用方式与上面的在线服务示例中的参数相同。下面展示了 choice 参数的一个用法示例

from vllm import LLM, SamplingParams
from vllm.sampling_params import GuidedDecodingParams

llm = LLM(model="HuggingFaceTB/SmolLM2-1.7B-Instruct")

guided_decoding_params = GuidedDecodingParams(choice=["Positive", "Negative"])
sampling_params = SamplingParams(guided_decoding=guided_decoding_params)
outputs = llm.generate(
    prompts="Classify this sentiment: vLLM is wonderful!",
    sampling_params=sampling_params,
)
print(outputs[0].outputs[0].text)

完整示例： examples/offline_inference/structured_outputs.py