Home › Python

ModuleNotFoundError: No module named 'langchain' (Python) — how to fix

Traceback (most recent call last):
  File "app.py", line 1, in <module>
    from langchain.chat_models import ChatOpenAI
ModuleNotFoundError: No module named 'langchain'

Quick fix: Install into the same interpreter: python -m pip install -U langchain. For integrations install the partner package too, e.g. python -m pip install langchain-openai and import from langchain_openai import ChatOpenAI. For No module named 'langchain_community' run pip install langchain-community.

What it means

Python searched sys.path of the current interpreter and found no langchain package. Either it's not installed there, or it's installed for a different Python (another venv, system Python, a different Jupyter kernel).

LangChain is also split into several distributions, so the module name in the error matters:

Error says Install
No module named 'langchain' langchain
No module named 'langchain_core' langchain-core
No module named 'langchain_community' langchain-community
No module named 'langchain_openai' langchain-openai
No module named 'langchain_anthropic' langchain-anthropic
No module named 'langchain_text_splitters' langchain-text-splitters

Note the pip name uses hyphens; the import name uses underscores.

Cause 1: not installed in this interpreter

Install with the exact Python you run:

python -m pip install -U langchain
python -c "import langchain, sys; print(langchain.__version__, sys.executable)"

Using python -m pip instead of bare pip avoids installing into a different Python. On Debian/Ubuntu system Python you'll get externally-managed-environment — use a venv:

python3 -m venv .venv
source .venv/bin/activate        # Windows: .venv\Scripts\activate
pip install langchain langchain-openai

Cause 2: Jupyter / VS Code uses another kernel

In a notebook, install into the kernel's own interpreter:

%pip install -U langchain langchain-openai

Then restart the kernel. In VS Code, pick the right interpreter with Python: Select Interpreter (the .venv one).

Cause 3: old import paths after upgrading

Many tutorials use imports from LangChain 0.0.x that were moved. Since 0.1/0.2 integrations live in separate packages:

# Old (fails or warns on current versions)
from langchain.chat_models import ChatOpenAI
from langchain.llms import OpenAI
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS
from langchain.document_loaders import PyPDFLoader

# Current
from langchain_openai import ChatOpenAI, OpenAI, OpenAIEmbeddings
from langchain_community.vectorstores import FAISS
from langchain_community.document_loaders import PyPDFLoader
from langchain_core.prompts import ChatPromptTemplate
pip install -U langchain langchain-openai langchain-community faiss-cpu pypdf

Cause 4: a file named langchain.py

If your script or a folder in the project is called langchain.py / langchain/, Python imports that instead and then can't find submodules (No module named 'langchain.chains'; 'langchain' is not a package). Rename it and delete __pycache__.

Cause 5: unsupported Python version

Current LangChain releases require Python 3.9+ (newer ones 3.10+). On older Python, pip may install an ancient version or fail with Could not find a version that satisfies the requirement. Check python --version and upgrade.

Pin versions

LangChain packages must be compatible with each other. Upgrade them together and pin in requirements.txt:

pip install -U langchain langchain-core langchain-community langchain-openai
pip freeze | grep -i langchain >> requirements.txt

Last updated 2026-11-19.