Use Mem0 storage

The mem0 backend uses Mem0 as long-term storage. It is a managed service where Mem0 handles memory extraction, storage, and retrieval, requiring no self-hosted vector store or local embedding — recommended for production.

When to use

Production needing a managed memory service;
You want Mem0 to automatically extract and manage memories;
You have a Mem0 account and API key, or can auto-fetch one via Volcengine credentials.

Dependencies

python -m pip install "veadk-python[database]"

Usage

The install command includes the official Mem0 Python package, mem0ai>=1.0.0,<2. Set DATABASE_MEM0_API_KEY and DATABASE_MEM0_BASE_URL first. Use https://api.mem0.ai for public Mem0 without a /v1 suffix. Volcengine credential exchange applies only to its corresponding managed memory service, not arbitrary Mem0 accounts

from veadk import Agent
from veadk.memory.long_term_memory import LongTermMemory

ltm = LongTermMemory(backend="mem0", app_name="ltm_demo")

agent = Agent(
    name="demo",
    instruction="Answer the user; when needed, use the `load_memory` tool to recall past conversations.",
    long_term_memory=ltm,
)

You can also pass config explicitly via backend_config:

import os

from veadk.memory.long_term_memory import LongTermMemory
from veadk.configs.database_configs import Mem0Config

ltm = LongTermMemory(
    backend="mem0",
    index="ltm_demo",
    backend_config={
        "index": "ltm_demo",
        "mem0_config": Mem0Config(
            api_key=os.environ["DATABASE_MEM0_API_KEY"],
            base_url="https://api.mem0.ai",
        ),
    },
)

Parameters

LongTermMemory parameters

Parameter	Type	Default	Description
`backend`	`str \| BaseLongTermMemoryBackend`	`"opensearch"`	Set to `mem0` for this page, or pass a configured backend instance
`backend_config`	`dict`	`{}`	Backend settings; a supplied backend instance takes precedence
`index`	`str`	`""`	Without backend_config, resolves from index, app_name, then default_app; supply a nonempty index with a configuration dictionary
`app_name`	`str`	`""`	Fallback for index; the actual user comes from the saved Session or search arguments
`top_k`	`int`	`5`	Number of retrieved chunks; use a positive integer
`user_id`	`str`	`""`	Deprecated; does not select the runtime user

Constructor parameters

backend_config supports the following settings:

Parameter	Type	Default	Description
`index`	`str`	No default; provided by `LongTermMemory`	Memory index name. The Mem0 backend imposes no naming constraints.
`mem0_config`	`Mem0Config`	Read automatically from `DATABASE_MEM0_*` env vars	Mem0 client config.

Mem0 config

mem0_config is a Mem0Config with env prefix DATABASE_MEM0_:

Field	Env var	Type	Default	Description
`api_key`	`DATABASE_MEM0_API_KEY`	`str`	`""`	Mem0 API key.
`api_key_id`	`DATABASE_MEM0_API_KEY_ID`	`str`	`""`	Used to auto-fetch the API key; provide this or `project_id`.
`project_id`	`DATABASE_MEM0_PROJECT_ID`	`str`	`""`	Used to auto-fetch the API key.
`base_url`	`DATABASE_MEM0_BASE_URL`	`str`	`""`	Mem0 service endpoint, e.g. `https://api.mem0.ai`.

API key resolution order

If api_key is provided directly, it is used as is; otherwise VeADK tries to fetch the API key via api_key_id or project_id; initialization fails if neither is set.

Environment variables

# Option 1: provide the API key directly
export DATABASE_MEM0_API_KEY="your-mem0-api-key"
export DATABASE_MEM0_BASE_URL="https://api.mem0.ai"

# Option 2: auto-fetch via api_key_id or project_id
export DATABASE_MEM0_API_KEY_ID="your-api-key-id"
export DATABASE_MEM0_PROJECT_ID="your-project-id"

If api_key is not provided directly, VeADK tries to obtain it via api_key_id or project_id; an error is raised if neither is set.

Verify writes and retrieval

After configuring the dependencies and credentials on this page, run this standalone example. It saves user text and searches for that user directly without calling a conversation model

import asyncio

from google.adk.events import Event
from google.adk.sessions import Session
from google.genai import types
from veadk.memory.long_term_memory import LongTermMemory

async def main():
    memory = LongTermMemory(backend="mem0", index="ltm_demo")
    session = Session(
        id="memory_check", app_name="ltm_demo", user_id="user_42",
        events=[Event(author="user", content=types.Content(
            role="user", parts=[types.Part(text="My preferred language is Chinese")]
        ))],
    )
    await memory.add_session_to_memory(session)
    result = await memory.search_memory(
        app_name="ltm_demo", user_id="user_42", query="preferred language"
    )
    for entry in result.memories:
        print(entry.content)

asyncio.run(main())

Results should contain the saved language preference. Managed services may extract memories asynchronously, so a completed write does not guarantee immediate retrieval. An empty result can also indicate permission, network, or service failure; check error logs and service records. The save method does not return a success Boolean index is not sent in Mem0 save or search requests and does not create a separate project. Within one service and credential scope, the runtime user_id distinguishes memory. Use application-specific credentials or stable application-and-user identifiers when sharing a service. Saving uses asynchronous extraction, so searches can initially return no results

Overview

Get started

Core components

Observability

Extensions

Frontend

Security

Deployment

When to use

Dependencies

Usage

Parameters

LongTermMemory parameters

Constructor parameters

Mem0 config

API key resolution order

Environment variables

Verify writes and retrieval

​When to use

​Dependencies

​Usage

​Parameters

​LongTermMemory parameters

​Constructor parameters

​Mem0 config

​API key resolution order

​Environment variables

​Verify writes and retrieval

When to use

Dependencies

Usage

Parameters

LongTermMemory parameters

Constructor parameters

Mem0 config

API key resolution order

Environment variables

Verify writes and retrieval