Skip to content
RememRemem Docs

Quickstart

  1. Get the Community release

    Remem 0.5 is a Go service backed by Pebble, an embeddable storage engine for Go. Clone the Community source snapshot and create its local environment file:

    Terminal window
    git clone https://github.com/remem-org/remem-community.git
    cd remem-community
    cp .env.example .env

  2. Set a server key and start Remem

    Replace the CHANGE_ME API key value in .env with a long random secret, then:

    Terminal window
    mkdir -p ./remem-data
    sudo chown 10001:10001 ./remem-data
    docker compose up --build -d

    Docker builds the Go server and its local embedding runtime. The embedding model is not part of the image: on the first start, a one-shot remem-model service downloads it (about 2.4 GB) into ./remem-model and checks every file, and the server starts once it has. Allow a few minutes for that first start; docker compose logs -f remem-model shows it arriving. Later starts find the model already verified and go straight on.

  3. Verify it’s running

    Terminal window
    curl http://localhost:4545/api/v1/ready

  4. Store your first memory

    Terminal window
    export REMEM_SERVER_API_KEY='your-secret-key'
    curl -X POST http://localhost:4545/api/v1/memories \
      -H "Authorization: Bearer $REMEM_SERVER_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "content": "User prefers concise bullet-point answers over long prose.",
        "policy": "long_term",
        "tags": ["preference", "communication"]
      }'

  5. Search memories

    Terminal window
    curl -X POST http://localhost:4545/api/v1/memories/search \
      -H "Authorization: Bearer $REMEM_SERVER_API_KEY" \
      -H "Content-Type: application/json" \
      -d '{"query":"how should I format my responses?","search_type":"hybrid","limit":5}'

    semantic finds related meaning, keyword matches terms, and hybrid combines both retrieval paths by rank.

Next steps

Section titled “Next steps”