Momento Cache is the world’s first truly serverless caching service, offering instant elasticity, scale-to-zero capability, and blazing-fast performance. Momento Vector Index stands out as the most productive, easiest-to-use, fully serverless vector index. For both services, simply grab the SDK, obtain an API key, input a few lines into your code, and you’re set to go. Together, they provide a comprehensive solution for your LLM data needs.
This page covers how to use the Momento ecosystem within LangChain.

​
Installation and Setup

  • Sign up for a free account here to get an API key
  • Install the Momento Python SDK with pip install momento

​
Cache

Use Momento as a serverless, distributed, low-latency cache for LLM prompts and responses. The standard cache is the primary use case for Momento users in any environment. To integrate Momento Cache into your application: 
from langchain.cache import MomentoCache
Then, set it up with the following code: 
from datetime import timedelta
from momento import CacheClient, Configurations, CredentialProvider
from langchain.globals import set_llm_cache

# Instantiate the Momento client
cache_client = CacheClient(
    Configurations.Laptop.v1(),
    CredentialProvider.from_environment_variable("MOMENTO_API_KEY"),
    default_ttl=timedelta(days=1))

# Choose a Momento cache name of your choice
cache_name = "langchain"

# Instantiate the LLM cache
set_llm_cache(MomentoCache(cache_client, cache_name))

​
Memory

Momento can be used as a distributed memory store for LLMs. See this notebook for a walkthrough of how to use Momento as a memory store for chat message history. 
from langchain.memory import MomentoChatMessageHistory

​
Vector Store

Momento Vector Index (MVI) can be used as a vector store. See this notebook for a walkthrough of how to use MVI as a vector store. 
from langchain_community.vectorstores import MomentoVectorIndex