Installation and setup

LangMem installs with pip and works with any LangChain chat model. The course uses a rule-based model, and a free Groq or Gemini key covers a real one.

pip install langmem 0.0.30

LangMem brings LangGraph and LangChain's core with it. The version is pinned because LangMem is an early release whose API may still change.

pip install "langmem==0.0.30"

Example

from importlib.metadata import version

print(version("langmem"))

A LangChain chat modelOptional

The course uses MemoryModel, a rule-based chat model you read in lesson 4, so no key is needed. Lesson 5 shows the change to a hosted model loaded with init_chat_model, which needs the provider's LangChain package and key.

python

from langchain.chat_models import init_chat_model

model = init_chat_model("google_genai:gemini-2.5-flash")

Provider	Key	Variable	Package and model string
Groq	Free, console.groq.com/keys	`GROQ_API_KEY`	`langchain-groq`, `"groq:openai/gpt-oss-120b"`
Gemini	Free, aistudio.google.com/apikey	`GOOGLE_API_KEY`	`langchain-google-genai`, `"google_genai:gemini-2.5-flash"`
OpenRouter	Free models end in `:free`, openrouter.ai/keys	`OPENROUTER_API_KEY`	`langchain-openrouter`, `"openrouter:google/gemma-4-31b-it:free"`
OpenAI	Paid, platform.openai.com/api-keys	`OPENAI_API_KEY`	`langchain-openai`, `"openai:gpt-4o-mini"`

pip install langchain-google-genai

export GOOGLE_API_KEY=AIza...

Try it yourself

Run pip show langmem and read which LangGraph version it requires.

PreviousLangMem overview

Little by little, you're building something great.