Repository files navigation

• Client Library Documentation
• Product Documentation

TheAlloyDB for PostgreSQL for LangChain package provides a first class experience for connecting toAlloyDB instances from the LangChain ecosystem while providing the following benefits:

• Simplified & Secure Connections: easily and securely create shared connection pools to connect to Google Cloud databases utilizing IAM for authorization and database authentication without needing to manage SSL certificates, configure firewall rules, or enable authorized networks.
• Improved performance & Simplified management: use a single-table schema can lead to faster query execution, especially for large collections.
• Improved metadata handling: store metadata in columns instead of JSON, resulting in significant performance improvements.
• Clear separation: clearly separate table and extension creation, allowing for distinct permissions and streamlined workflows.
• Better integration with AlloyDB: built-in methods to take advantage of AlloyDB's advanced indexing and scalability capabilities.

In order to use this library, you first need to go through the followingsteps:

1. Select or create a Cloud Platform project.
2. Enable billing for your project.
3. Enable the AlloyDB API.
4. Setup Authentication.

Install this library in avirtualenv using pip.virtualenv is a tool to create isolated Python environments. The basic problem it addresses isone of dependencies and versions, and indirectly permissions.

Withvirtualenv, it'spossible to install this library without needing system installpermissions, and without clashing with the installed systemdependencies.

Python >= 3.9

pip install virtualenvvirtualenv <your-env>source <your-env>/bin/activate<your-env>/bin/pip install langchain-google-alloydb-pg

pip install virtualenvvirtualenv <your-env><your-env>\Scripts\activate<your-env>\Scripts\pip.exe install langchain-google-alloydb-pg

Use a vector store to store embedded data and perform vector search.

fromlangchain_google_alloydb_pgimportAlloyDBEngine,AlloyDBVectorStorefromlangchain_google_vertexaiimportVertexAIEmbeddingsengine=AlloyDBEngine.from_instance("project-id","region","my-cluster","my-instance","my-database")embeddings_service=VertexAIEmbeddings(model_name="textembedding-gecko@003")vectorstore=AlloyDBVectorStore.create_sync(engine,table_name="my-table",embedding_service=embeddings_service)

See the fullVector Store tutorial.

Use a document loader to load data as LangChainDocuments.

fromlangchain_google_alloydb_pgimportAlloyDBEngine,AlloyDBLoaderengine=AlloyDBEngine.from_instance("project-id","region","my-cluster","my-instance","my-database")loader=AlloyDBLoader.create_sync(engine,table_name="my-table-name")docs=loader.lazy_load()

See the fullDocument Loader tutorial.

UseChatMessageHistory to store messages and provide conversationhistory to LLMs.

fromlangchain_google_alloydb_pgimportAlloyDBChatMessageHistory,AlloyDBEngineengine=AlloyDBEngine.from_instance("project-id","region","my-cluster","my-instance","my-database")history=AlloyDBChatMessageHistory.create_sync(engine,table_name="my-message-store",session_id="my-session-id")

See the fullChat Message History tutorial.

UseAlloyDBSaver to save snapshots of the graph state at a given point in time.

fromlangchain_google_alloydb_pgimportAlloyDBSaver,AlloyDBEngineengine=AlloyDBEngine.from_instance("project-id","region","my-cluster","my-instance","my-database")checkpoint=AlloyDBSaver.create_sync(engine)

See the fullCheckpoint tutorial.

Code examples can be found in thesamples/ folder.

Async functionality improves the speed and efficiency of database connections through concurrency,which is key for providing enterprise quality performance and scaling in GenAI applications. Thispackage uses a native async Postgres driver,asyncpg, to optimize Python's async functionality.

LangChain supportsasync programming, since LLM based application utilize many I/O-bound operations,such as making API calls to language models, databases, or other services. All components should provideboth async and sync versions of all methods.

asyncio is a Python library used for concurrent programming and is used as the foundation for multiplePython asynchronous frameworks. asyncio uses async / await syntax to achieve concurrency fornon-blocking I/O-bound tasks using one thread with cooperative multitasking instead of multi-threading.

Update sync methods to await async methods

engine=awaitAlloyDBEngine.afrom_instance("project-id","region","my-cluster","my-instance","my-database")awaitengine.ainit_vectorstore_table(table_name="my-table",vector_size=768)vectorstore=awaitAlloyDBVectorStore.create(engine,table_name="my-table",embedding_service=VertexAIEmbeddings(model_name="textembedding-gecko@003"))

ipython and jupyter notebooks support the use of the await keyword without any additional setup

Update routes to use async def.

@app.get("/invoke/")asyncdefinvoke(query:str):returnawaitretriever.ainvoke(query)

It is recommend to create a top-level async method definition: async def to wrap multiple async methods.Then use asyncio.run() to run the the top-level entrypoint, e.g. "main()"

asyncdefmain():response=awaitretriever.ainvoke(query)print(response)asyncio.run(main())

Contributions to this library are always welcome and highly encouraged.

SeeCONTRIBUTING andDEVELOPER for more information how to get started.

Please note that this project is released with a Contributor Code of Conduct. By participating inthis project you agree to abide by its terms. SeeCode of Conduct for moreinformation.

Apache 2.0 - SeeLICENSEfor more information.

This is not an officially supported Google product.