I added self-hostable memory backend support for google adk using openmemory

[rakshith-git]

Hey everyone! Just finished my first major contribution pull request to Google's ADK and wanted to share.

What I built: Self-hosted memory backend support using OpenMemory - basically giving AI agents long-term memory without needing cloud services and it works exactly like Vertex AI memory bank.

ADK only supported Vertex AI memory before, which meant you needed Google Cloud to give your agents memory. Now you can run everything locally or on your own infrastructure.

Here's the usage - super simple:

from google.adk import Agent, Runner
from google.adk.memory import OpenMemoryService

memory = OpenMemoryService(base_url="http://localhost:3000")

agent = Agent(
    name="my_agent",
    model="gemini-2.0-flash",
    instruction="You remember past conversations."
)

runner = Runner(agent=agent, memory_service=memory)

# Now your agent remembers across sessions
await runner.run("My favorite color is blue")
# Later in a new session...
await runner.run("What's my favorite color?")  # "blue" ✅

Or just use the CLI:

adk web agents_dir --memory_service_uri="openmemory://localhost:3000"

Cool features:

Multi-sector embeddings (brain-inspired memory organization)

Memory decay over time

Multi-user support with server-side filtering

Runs completely self-hosted

Install:

pip install google-adk[openmemory]

This is my first big open source contribution so any feedback would be awesome