omemo

@OMNIDIMEN

Add a layer of replaceable long-term memory between any OpenAI compatible client and the model.

omemo sits between the chat application and the upstream model API, saving conversations, extracting memories and automatically injecting them in the next round. The client only needs to point the Base URL to the proxy to continue using the original model and interface.

Project address (can be copied to AI):https://github.com/OmniDimen/omemo

PROJECT INTRO

Project introduction

omemo is an OpenAI compatible memory agent. The chat client sends the request to omemo, and the agent saves the conversation, calls the built-in or external summary mode to form a memory, and then adds the relevant content to subsequent model requests.

The project supports two routes: complete prompt injection and RAG, which is suitable for users who do not want to change the client and only want to adjust the API address.

Suitable for:There are already OpenAI compatible clients for those who want to gain long-term memory with minimal changes.
Platform:Python Self-host, connect to any OpenAI compatible application and model service.
Usage:Deploy the agent, configure the upstream model, and change the client Base URL to omemo.
License:Apache-2.0。

◌

COMMENTS & FEEDBACK

Comments and Feedback

Guest message · Will be published after review

People who have used this project can come back and tell newcomers: which platform it ran on, whether there were any pitfalls during the installation, and what the actual experience was like. The current version is in guest mode, and you do not need to register an account to leave a message.

Reading public comments...

SOURCE & CREDIT

Atlas is responsible for the introduction and navigation, and the use still returns to the original author.

We do not mirror project files, nor do we intercept author traffic. For installation, download, version updates and the latest instructions, please refer to the project page provided by the original author.

Project link ↗