LangChain Classic

1.0.3 · deprecated · verified Sun Apr 05

LangChain Classic is a legacy Python package providing the original API for building applications with Large Language Models (LLMs) through composability. It represents the state of the `langchain` library prior to its major architectural refactor (pre-0.1.0/0.2.0, before v1.0), which split the project into `langchain-core`, `langchain-community`, and partner packages. This package is in maintenance mode for backward compatibility and does not receive new features; it is not recommended for new projects.

Warnings

breaking LangChain Classic's API is fundamentally incompatible with the modern LangChain (v1.0+) ecosystem, including `langchain-core`, `langchain-community`, and partner packages. Direct migration requires significant code changes.
Fix: For new projects, use `langchain` (v1.0+) and its modular components (`langchain-core`, `langchain-openai`, etc.). For existing projects, consider a full migration to the new API or continue using `langchain-classic` in isolation.
deprecated The `langchain-classic` package is explicitly for backward compatibility with pre-v1.0 LangChain and is not actively developed for new features.
Fix: New development should target the latest `langchain` package (v1.0+) and its structured approach to agents, models, and tools. Refer to the official LangChain migration guides.
gotcha Attempting to mix `langchain-classic` with newer `langchain` packages (e.g., `langchain-core`, `langchain-community`, `langchain-openai`) will likely lead to dependency conflicts, import errors, and runtime issues due to incompatible APIs and overlapping namespaces.
Fix: Maintain strict separation. If using `langchain-classic`, avoid installing or using any other `langchain-*` packages that are part of the v1.0+ ecosystem, unless explicitly stated in `langchain-classic` documentation as compatible dependencies. Check `pip freeze` for installed `langchain` related packages.
gotcha LangChain Classic will not receive new features, performance optimizations, or regular bug fixes available in the ongoing development of the main `langchain` library. This includes updates for new LLM capabilities or provider integrations.
Fix: To access the latest advancements in LLM technology and LangChain's framework, a migration to the modern `langchain` architecture is necessary.

Install

pip install langchain-classic Install LangChain Classic

Imports

LLMChain
```
from langchain.chains import LLMChain
```
Legacy chains are directly imported from `langchain.chains` in classic. In modern LangChain, these are often in `langchain_community` or replaced by new patterns.
OpenAI
```
from langchain.llms import OpenAI
```
LLM integrations were directly under `langchain.llms` or `langchain.chat_models`. In modern LangChain, specific provider integrations moved to separate packages like `langchain_openai`.
PromptTemplate
```
from langchain.prompts import PromptTemplate
```
Prompt templates were directly under `langchain.prompts`. In modern LangChain, core prompt components moved to `langchain_core.prompts`.

Quickstart

This quickstart demonstrates a basic LLM interaction using an `LLMChain` with an OpenAI model and a `PromptTemplate`, typical of `langchain-classic` usage. Ensure `OPENAI_API_KEY` is set in your environment variables.

import os
from langchain.llms import OpenAI
from langchain.prompts import PromptTemplate
from langchain.chains import LLMChain

# Set your OpenAI API key as an environment variable (e.g., export OPENAI_API_KEY="sk-...")
# For this example, we use os.environ.get to prevent errors if the key is not set.
openai_api_key = os.environ.get('OPENAI_API_KEY', '')

if not openai_api_key:
    print("Warning: OPENAI_API_KEY environment variable not set. Please set it to run the example.")
    print("Skipping LLM interaction.")
else:
    llm = OpenAI(openai_api_key=openai_api_key, temperature=0.7)

    prompt = PromptTemplate(
        input_variables=["topic"],
        template="Tell me a short, funny joke about {topic}."
    )

    chain = LLMChain(llm=llm, prompt=prompt)

    try:
        response = chain.run("developers")
        print(f"Joke about developers:\n{response}")
    except Exception as e:
        print(f"An error occurred during LLM interaction: {e}")
        print("Please ensure your OPENAI_API_KEY is valid and has access to the OpenAI API.")

view raw JSON →