https://github.com/memodb-io/memobase
Profile-Based Long-Term Memory for AI Applications
https://github.com/memodb-io/memobase
ai-companion ai-memory chatgpt data-structures data-structures-and-algorithms llm-application llm-memory long-term-memory memory rag retrieval user-memory
Last synced: 2 days ago
JSON representation
Profile-Based Long-Term Memory for AI Applications
- Host: GitHub
- URL: https://github.com/memodb-io/memobase
- Owner: memodb-io
- License: apache-2.0
- Created: 2024-09-03T03:20:53.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2025-04-12T02:20:52.000Z (3 days ago)
- Last Synced: 2025-04-12T21:18:45.338Z (2 days ago)
- Topics: ai-companion, ai-memory, chatgpt, data-structures, data-structures-and-algorithms, llm-application, llm-memory, long-term-memory, memory, rag, retrieval, user-memory
- Language: Python
- Homepage: https://memobase.io
- Size: 8.15 MB
- Stars: 1,027
- Watchers: 6
- Forks: 67
- Open Issues: 6
-
Metadata Files:
- Readme: readme.md
- Changelog: Changelog.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Codeowners: .github/CODEOWNERS
Awesome Lists containing this project
- awesome-ChatGPT-repositories - memobase - Profile-Based Long-Term Memory for AI Applications (Others)
- awesome-hacking-lists - memodb-io/memobase - Profile-Based Long-Term Memory for AI Applications (Python)
README
Memobase
Memobase is a **user profile-based memory system** designed to bring long-term user memory to your Generative AI (GenAI) applications. Whether you're building virtual companions, educational tools, or personalized assistants, Memobase empowers your AI to **remember**, **understand**, and **evolve** with your users.
Memobase can provide you structured profiles of users, check out the [result](./docs/experiments/900-chats/readme.md) (compared with [mem0](https://github.com/mem0ai/mem0)) from a 900-turns real-world chatting:
Partial Profile Output
```python
{
"basic_info": {
"language_spoken": ["English", "Korean"],
"name": "오*영"
},
"demographics": {
"marital_status": "married"
},
"education": {
"notes": "Had an English teacher who emphasized capitalization rules during school days",
"major": "국어국문학과 (Korean Language and Literature)"
},
"interest": {
"games": 'User is interested in Cyberpunk 2077 and wants to create a game better than it',
'youtube_channels': "Kurzgesagt",
...
},
"psychological": {...},
'work': {'working_industry': ..., 'title': ..., },
...
}
```
## Core Features
- **🎯 Memory for User, not Agent**: Define and control exactly what user information your AI captures.
- ➡️ **Time-aware Memory**: Memobase saves specific dates in profiles to prevent outdated information from affecting your AI. Also, check [Memobase event](https://docs.memobase.io/features/event/event) for sequential events (episodic memory).
- **🖼️ Contorllable Memory**: Among all types of memory, only some may enhance your product experience. Memobase offers a flexible configuration for you to [design the profile](https://docs.memobase.io/features/profile/profile).
- **🔌 Easy Integration**: Minimal code changes to integrate with your existing LLM stack with [API](https://docs.memobase.io/api-reference/overview), [Python](https://pypi.org/project/memobase/)/[Node](./src/client/memobase-ts/README.md)/[Go](./src/client/memobase-go/README.md) SDK.
- **⚡️ Insert with Buffer**: Memory system will cost you extra money, Memobase offers every user a buffer to batch processing the chats after the conversation. Fast & Cheap.
- **🚀 Production Ready**: Memobase is building with FastAPI, Postgres and Redis, supporting request caching, authing, telemetry... [Fully dockerized](./src/server/readme.md).
How Memobase works?
## Get Started
1. [Start your Memobase server locally](./src/server/readme.md). If you don't want to be bothered, Memobase Cloud provides [a free tier](https://www.memobase.io/en/login) enough for your testing
2. You should have the below two things to continue:
1. A project url. (local: `http://localhost:8019` , cloud `https://api.memobase.dev`)
2. A project token. (local: `secret` , cloud `sk-proj-xxxxxx`)
3. Install the Python SDK: `pip install memobase`
4. Below tutorial is for Python User. For other language and API, check [this](https://docs.memobase.io/quickstart).
## Step-by-step break down
> [!TIP]
>
> You can just run this equivalent [quickstart script](./assets/quickstart.py)
>
> Or you can keep things super easy by using [OpenAI SDK with Memobase.](https://docs.memobase.io/practices/openai), [Ollama with Memobase](./assets/tutorials/ollama+memobase)
### 1. Make sure you're connected
```python
from memobase import MemoBaseClient, ChatBlob
mb = MemoBaseClient(
project_url=PROJECT_URL,
api_key=PROJECT_TOKEN,
)
assert mb.ping()
```
### 2. Manage Users
```python
uid = mb.add_user({"any_key": "any_value"})
mb.update_user(uid, {"any_key": "any_value2"})
u = mb.get_user(uid)
print(u)
# mb.delete(uid)
```
### 3. Insert Data
> In Memobase, all types of data are blobs to a user that can insert, get and delete:
```python
messages = [
{
"role": "user",
"content": "Hello, I'm Gus",
},
{
"role": "assistant",
"content": "Hi, nice to meet you, Gus!",
}
]
bid = u.insert(ChatBlob(messages=messages))
print(u.get(bid)) # not found once you flush the memory.
# u.delete(bid)
```
> Be default, Memobase will remove the blobs once they're processed. This means that apart from the relevant memory, your data will not be stored with Memobase. You can persist the blobs by adjusting the [configuration file](https://docs.memobase.io/features/customization/full#storage-config).
### 4. Get your Memory
```python
u.flush()
```
And what will you get?
```python
print(u.profile(need_json=True))
# results
{
"basic_info": {
"name": {
"content": "Gus",
"id": ...,
"created_at": ...
}
}
}
```
`u.profile()` will return structured profiles that are learned from this user, including `topic`, `sub_topic` and `content`. As you insert more blobs, the profile will become better.
Why need a flush?
In Memobase, we don't memoize users in [hot path](https://langchain-ai.github.io/langgraph/concepts/memory/#writing-memories-in-the-hot-path). We use buffer zones for the recent inserted blobs.
When the buffer zone becomes too large (e.g., 1024 tokens) or remains idle for an extended period (e.g., 1 hour), Memobase will flush the entire buffer into memory. Alternatively, you can use `flush()` manually decide when to flush, such as when a chat session is closed in your app.
### 5. Integrate memory into your prompt
Memobase has a `context` api to pack everything you need into a simple string, where you can insert it into your prompt directly:
```python
print(u.context(max_token_size=500, prefer_topics=["basic_info"]))
```
Something like:
```
# Below is the user profile:
- basic_info::name: Gus
...
# Below is the latest events of the user:
2025/02/24 04:25PM:
- work::meetings: Scheduled a meeting with John.
...
Please provide your answer using the information within the tag at the appropriate time.
```
Checkout the detail params [here](https://docs.memobase.io/api-reference/prompt/get_context).
### What's next?
- Checkout the [quickstart script](./assets/quickstart.py) for more details
- You may want to explore the [customization](https://docs.memobase.io/features/profile/profile) of Memobase to make sure the system works as your expectation.
- If you want to test Memobase on your own data, we offer a [script](./docs/experiments/chat_sessions) that allows you to set multiple chat sessions and see how the memory grows.
## Why/Where should I use Memobase?
### Remember the users
By placing profiles into your AI (*e.g.* system prompt).
Demo
```python
PROFILES = "\n".join([p.describe for p in u.profile()])
print(PROFILES)
# basic_info: name - Gus
# basic_info: age - 25
# ...
# interest: foods - Mexican cuisine
# psychological: goals - Build something that maybe useful
# ...
```
### User analysis and tracking
Too much information is hidden in the conversations between users and AI, that's why you need a new data tracking method to record user preference and behavior.
Demo
```python
PROFILES = u.profile()
def under_age_30(p):
return p.sub_topic == "age" and int(p.content) < 30
def love_cat(p):
return p.topic == "interest" and p.sub_topic == "pets" and "cat" in p.content
is_user_under_30 = (
len([p for p in profiles if under_age_30(p)]) > 0
)
is_user_love_cat = (
len([p for p in profiles if love_cat(p)]) > 0
)
...
```
### Sell something to your customers.
Not everyone is looking for Grammarly, it's always nice to sell something your users might want.
Demo
```python
def pick_an_ad(profiles):
work_titles = [p for p in profiles if p.topic=="work" and p.sub_topic=="title"]
if not len(work_titles):
return None
wt = work_titles[0].content
if wt == "Software Engineer":
return "Deep Learning Stuff"
elif wt == "some job":
return "some ads"
...
```
## Documentation
For detailed usage instructions, visit the [documentation](https://docs.memobase.io/).
## Stay Updated
Star Memobase on Github to support and receive instant notifications!
## Support
Join the community for support and discussions:
- [Join our Discord](https://discord.gg/YdgwU4d9NB) 👻
- [Follow us on Twitter](https://x.com/memobase_io) 𝕏
Or Just [email us](mailto:[email protected]) ❤️
## Contribute
- Check out our [Changelog](./Changelog.md) first, make sure the feature you want has not been developed or is currently being planned.:)
- Go through [Contributing](./CONTRIBUTING.md) document to setup and contribute to Memobase.
## License
This project is licensed under the Apache 2.0 License - see the [LICENSE](https://github.com/memodb-io/memobase/blob/main/LICENSE) file for details.