AR BG BN CS DA DE EL EN ES FA FI FR HE HI HR HU ID IT JA KO LT MS NL NO PL PT RO RU SK SV SW TA TH TL TR UK UR VI ZH-TW ZH

LangChain 集成指南

了解如何将 LangChain 与 AI Gateway 集成，统一访问多个 AI 模型。

LangChain 集成指南

LangChain 是构建大语言模型应用最流行的框架之一。AI Gateway 与 LangChain 无缝集成，让您能够通过单个 API 统一访问 100 多个模型。

Compatibility

AI Gateway is fully compatible with LangChain's OpenAI integration. You can swap providers in just a few lines of code.

前提条件

Python 3.8 或更高版本
LangChain 0.1.0 或更高版本
AI Gateway API 密钥

安装

Install LangChain and OpenAI packages

bash

pip install langchain-openai langchain-core

Set up your API key

Set your AI Gateway API key as an environment variable:

python

import os
os.environ["OPENAI_API_KEY"] = "your-ai-gateway-key"
os.environ["OPENAI_API_BASE"] = "https://api.ecaix.com/v1"

Initialize the chat model

Now you can use any model with LangChain:

python

from langchain_openai import ChatOpenAI

# Use any model - just change the model name
llm = ChatOpenAI(
    model="gpt-4o",
    temperature=0.7,
)

response = llm.invoke("Hello, world!")
print(response.content)

Using Multiple Models

One of the key benefits of using AI Gateway with LangChain is easy model switching:

python

from langchain_openai import ChatOpenAI

# Define models to compare
models = ["gpt-4o", "claude-3-5-sonnet", "deepseek-v3"]

for model_name in models:
    llm = ChatOpenAI(model=model_name, temperature=0.7)
    response = llm.invoke("Explain quantum computing in one sentence.")
    print(f"{model_name}: {response.content}\n")

Using with LangChain Expression Language (LCEL)

AI Gateway works perfectly with LangChain's powerful chaining capabilities:

python

from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser

# Create a simple chain
prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a helpful assistant that speaks {language}."),
    ("human", "{question}")
])

model = ChatOpenAI(model="gpt-4o")
output_parser = StrOutputParser()

chain = prompt | model | output_parser

# Invoke the chain
result = chain.invoke({
    "language": "Spanish",
    "question": "What is AI?"
})
print(result)

Streaming Responses

For better user experience, use streaming:

python

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(model="gpt-4o", streaming=True)

for chunk in llm.stream("Write a haiku about AI:"):
    print(chunk.content, end="", flush=True)

Using with Agents

AI Gateway works with LangChain Agents for complex task automation:

python

from langchain_openai import ChatOpenAI
from langchain.agents import AgentExecutor, create_react_agent
from langchain_core.tools import tool
from langchain import hub

# Define a simple tool
@tool
def get_weather(city: str) -> str:
    """Get the weather for a city."""
    return f"The weather in {city} is sunny."

# Create the agent
llm = ChatOpenAI(model="gpt-4o")
tools = [get_weather]
agent = create_react_agent(llm, tools, hub.pull("hwchase17/react"))

# Run the agent
agent_executor = AgentExecutor(agent=agent, tools=tools)
result = agent_executor.invoke({
    "input": "What's the weather in San Francisco?"
})
print(result["output"])

Advanced: Custom Model Selection

Use AI Gateway's automatic routing to select the best model for each task:

python

from langchain_openai import ChatOpenAI

# AI Gateway can automatically route requests based on task complexity
# Simple tasks → Fast/cheap models
# Complex tasks → Powerful models

llm = ChatOpenAI(
    model="auto",  # AI Gateway handles routing automatically
    temperature=0.7,
)

# The model parameter "auto" tells AI Gateway to:
# - Analyze the request complexity
# - Select the optimal model
# - Balance cost and performance
response = llm.invoke("Write a complex Python decorator.")
print(response.content)

Best Practice

Use the "auto" model routing for production applications to optimize cost-performance balance automatically.

Next Steps

← Back to Integrations

AR BG BN CS DA DE EL EN ES FA FI FR HE HI HR HU ID IT JA KO LT MS NL NO PL PT RO RU SK SV SW TA TH TL TR UK UR VI ZH-TW ZH

LangChain 集成指南

了解如何将 LangChain 与 AI Gateway 集成，统一访问多个 AI 模型。

LangChain 集成指南

LangChain 是构建大语言模型应用最流行的框架之一。AI Gateway 与 LangChain 无缝集成，让您能够通过单个 API 统一访问 100 多个模型。

Compatibility

AI Gateway is fully compatible with LangChain's OpenAI integration. You can swap providers in just a few lines of code.

前提条件

Python 3.8 或更高版本
LangChain 0.1.0 或更高版本
AI Gateway API 密钥

安装

Install LangChain and OpenAI packages

bash

pip install langchain-openai langchain-core

Set up your API key

Set your AI Gateway API key as an environment variable:

python

import os
os.environ["OPENAI_API_KEY"] = "your-ai-gateway-key"
os.environ["OPENAI_API_BASE"] = "https://api.ecaix.com/v1"

Initialize the chat model

Now you can use any model with LangChain:

python

from langchain_openai import ChatOpenAI

# Use any model - just change the model name
llm = ChatOpenAI(
    model="gpt-4o",
    temperature=0.7,
)

response = llm.invoke("Hello, world!")
print(response.content)

Using Multiple Models

One of the key benefits of using AI Gateway with LangChain is easy model switching:

python

from langchain_openai import ChatOpenAI

# Define models to compare
models = ["gpt-4o", "claude-3-5-sonnet", "deepseek-v3"]

for model_name in models:
    llm = ChatOpenAI(model=model_name, temperature=0.7)
    response = llm.invoke("Explain quantum computing in one sentence.")
    print(f"{model_name}: {response.content}\n")

Using with LangChain Expression Language (LCEL)

AI Gateway works perfectly with LangChain's powerful chaining capabilities:

python

from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser

# Create a simple chain
prompt = ChatPromptTemplate.from_messages([
    ("system", "You are a helpful assistant that speaks {language}."),
    ("human", "{question}")
])

model = ChatOpenAI(model="gpt-4o")
output_parser = StrOutputParser()

chain = prompt | model | output_parser

# Invoke the chain
result = chain.invoke({
    "language": "Spanish",
    "question": "What is AI?"
})
print(result)

Streaming Responses

For better user experience, use streaming:

python

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(model="gpt-4o", streaming=True)

for chunk in llm.stream("Write a haiku about AI:"):
    print(chunk.content, end="", flush=True)

Using with Agents

AI Gateway works with LangChain Agents for complex task automation:

python

from langchain_openai import ChatOpenAI
from langchain.agents import AgentExecutor, create_react_agent
from langchain_core.tools import tool
from langchain import hub

# Define a simple tool
@tool
def get_weather(city: str) -> str:
    """Get the weather for a city."""
    return f"The weather in {city} is sunny."

# Create the agent
llm = ChatOpenAI(model="gpt-4o")
tools = [get_weather]
agent = create_react_agent(llm, tools, hub.pull("hwchase17/react"))

# Run the agent
agent_executor = AgentExecutor(agent=agent, tools=tools)
result = agent_executor.invoke({
    "input": "What's the weather in San Francisco?"
})
print(result["output"])

Advanced: Custom Model Selection

Use AI Gateway's automatic routing to select the best model for each task:

python

from langchain_openai import ChatOpenAI

# AI Gateway can automatically route requests based on task complexity
# Simple tasks → Fast/cheap models
# Complex tasks → Powerful models

llm = ChatOpenAI(
    model="auto",  # AI Gateway handles routing automatically
    temperature=0.7,
)

# The model parameter "auto" tells AI Gateway to:
# - Analyze the request complexity
# - Select the optimal model
# - Balance cost and performance
response = llm.invoke("Write a complex Python decorator.")
print(response.content)

Best Practice

Use the "auto" model routing for production applications to optimize cost-performance balance automatically.

Next Steps

← Back to Integrations