longkeyy/fai-memory repository overview

FAI Memory Service is built for Large Language Models (LLMs) to store, retrieve, and manage conversational memory at scale.
Using Docker, you can quickly deploy in a consistent, isolated, and scalable environment.

Benefits:

• Optimized for LLMs – efficient handling of conversation context
• Multiple backends – Redis, MongoDB, SQLite, or in-memory
• Easy deployment – single command setup
• Horizontal scaling for high availability

⁠Run Container

docker run -d   --name fai-memory   -p 8080:8080   -v /path/to/config:/app/conf   fai-memory

Key Options:

• -p 8080:8080: Map API endpoint to host port
• -v /path/to/config:/app/conf: Mount custom configuration for memory backend

⁠Verify Service Health

curl http://localhost:8080/health

Expected output:

{
  "status": "ok",
  "message": "Service is healthy"
}

⁠Docker Compose Example

For production or multi-service setups:

version: '3.9'
services:
  fai-memory:
    image: fai-memory:latest
    ports:
      - "8080:8080"
    volumes:
      - ./config:/app/conf
    depends_on:
      - redis

  redis:
    image: redis:6.2
    ports:
      - "6379:6379"

Start with:

docker-compose up -d

⁠Summary

Docker makes it simple to deploy FAI Memory Service for LLMs, providing a scalable and reliable memory layer to support complex conversational AI systems.