⚡ Introducing CORE - open source, shareable, user-owned memory graph for LLMs

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • MyrinNew
    Senior Member
    • Feb 2024
    • 5175

    #1

    ⚡ Introducing CORE - open source, shareable, user-owned memory graph for LLMs

    Problem

    Are you wasting time re-explaining yourself dozens of times a day, just to keep ChatGPT, Cursor, and Claude on the same page?


    ChatGPT might recall your project details, but Cursor forgets them, and Claude starts from scratch every time.


    Meet CORE

    An open source, shareable knowledge graph (your memory vault) that lets any LLM (ChatGPT, Cursor, Claude, SOL, etc.) share and query the same persistent context.







    CORE is

    ✅ Shareable

    ✅ Relational - Every fact gets a full version history (who, when, why)

    ✅ 100% owned by you: Your memory data, you decide what to store and retrieve.

    Get Started with Core

    CORE Cloud

    You can try for free (limited period) on https://core.heysol.ai


    Run Locally


    STEP 1 - Copy the example environment file to .env:






    cp .env.example .env







    STEP 2 - Use Docker Compose to start all required services:






    docker-compose up







    STEP 3 - Access the app

    Once the containers are running, open your browser and go to http://localhost:3000.


    STEP 4 - Login with Magic Link
    • Choose the "Magic Link" login option.
    • Enter your email.
    • Copy the magic link from terminal logs and open it in your browser.


    STEP 5 - Create Your Private Space & Ingest Data
    • In the dashboard, go to the ingest section.
    • Type a message, e.g., I love playing badminton, and click "Add".
    • Your memory is queued for processing; you can monitor its status in - the server logs.
    • Once processing is complete, nodes will be added to your private knowledge graph and visible in the dashboard.
    • You can later choose to connect this memory to other tools or keep it private.


    STEP 6 - Search Your Memory

    Use the dashboard's search feature to query your ingested data within your private space.


    Note: We are actively working on improving support for Llama models. At the moment, C.O.R.E does not provide optimal results with Llama-based models, but we are making progress to ensure better compatibility and output in the near future.


    For more information, visit the official docs, and their website.


    Star the CORE repository ⭐





    More...
Working...