从LLMRouterChain迁移

通过 LLMRouterChain，输入查询被路由到一个或多个目标——即，给定一个输入查询，它使用 LLM 从目标链列表中进行选择，并将其输入传递给选定的链。

LLMRouterChain 不支持常见的聊天模型功能，例如消息角色和工具调用。在底层，LLMRouterChain 通过指示 LLM 生成 JSON 格式的文本并解析出预期的目标来路由查询。

考虑一个来自MultiPromptChain的例子，它使用了LLMRouterChain。下面是一个（示例）默认提示

from langchain.chains.router.multi_prompt import MULTI_PROMPT_ROUTER_TEMPLATE

destinations = """
animals: prompt for animal expert
vegetables: prompt for a vegetable expert
"""

router_template = MULTI_PROMPT_ROUTER_TEMPLATE.format(destinations=destinations)

print(router_template.replace("`", "'"))  # for rendering purposes

Given a raw text input to a language model select the model prompt best suited for the input. You will be given the names of the available prompts and a description of what the prompt is best suited for. You may also revise the original input if you think that revising it will ultimately lead to a better response from the language model.

<< FORMATTING >>
Return a markdown code snippet with a JSON object formatted to look like:
'''json
{{
    "destination": string \ name of the prompt to use or "DEFAULT"
    "next_inputs": string \ a potentially modified version of the original input
}}
'''

REMEMBER: "destination" MUST be one of the candidate prompt names specified below OR it can be "DEFAULT" if the input is not well suited for any of the candidate prompts.
REMEMBER: "next_inputs" can just be the original input if you don't think any modifications are needed.

<< CANDIDATE PROMPTS >>

animals: prompt for animal expert
vegetables: prompt for a vegetable expert


<< INPUT >>
{input}

<< OUTPUT (must include '''json at the start of the response) >>
<< OUTPUT (must end with ''') >>

大部分行为是通过单个自然语言提示确定的。支持工具调用功能的聊天模型为这项任务带来了许多优势：

• 支持聊天提示模板，包括带有system及其他角色的消息；
• 工具调用模型经过微调以生成结构化输出；
• 支持可运行方法，例如流式传输和异步操作。

现在我们来比较LLMRouterChain与使用工具调用的 LCEL 实现。请注意，本指南将使用langchain-openai >= 0.1.20

%pip install -qU langchain-core langchain-openai

import os
from getpass import getpass

if "OPENAI_API_KEY" not in os.environ:
    os.environ["OPENAI_API_KEY"] = getpass()

旧版

详情

from langchain.chains.router.llm_router import LLMRouterChain, RouterOutputParser
from langchain_core.prompts import PromptTemplate
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(model="gpt-4o-mini")

router_prompt = PromptTemplate(
    # Note: here we use the prompt template from above. Generally this would need
    # to be customized.
    template=router_template,
    input_variables=["input"],
    output_parser=RouterOutputParser(),
)

chain = LLMRouterChain.from_llm(llm, router_prompt)

API 参考：LLMRouterChain | RouterOutputParser | PromptTemplate | ChatOpenAI

result = chain.invoke({"input": "What color are carrots?"})

print(result["destination"])

vegetables

LCEL

详情

from operator import itemgetter
from typing import Literal

from langchain_core.prompts import ChatPromptTemplate
from langchain_core.runnables import RunnablePassthrough
from langchain_openai import ChatOpenAI
from typing_extensions import TypedDict

llm = ChatOpenAI(model="gpt-4o-mini")

route_system = "Route the user's query to either the animal or vegetable expert."
route_prompt = ChatPromptTemplate.from_messages(
    [
        ("system", route_system),
        ("human", "{input}"),
    ]
)


# Define schema for output:
class RouteQuery(TypedDict):
    """Route query to destination expert."""

    destination: Literal["animal", "vegetable"]


# Instead of writing formatting instructions into the prompt, we
# leverage .with_structured_output to coerce the output into a simple
# schema.
chain = route_prompt | llm.with_structured_output(RouteQuery)

API 参考：ChatPromptTemplate | RunnablePassthrough | ChatOpenAI

result = chain.invoke({"input": "What color are carrots?"})

print(result["destination"])

vegetable

下一步

有关使用提示模板、LLM 和输出解析器进行构建的更多详细信息，请参阅本教程。

有关更多背景信息，请查看LCEL 概念文档。