> ## Documentation Index
> Fetch the complete documentation index at: https://docs.orcamemory.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Requirements

> What you need to self-host Orca Memory

## Overview

Orca Memory is built on modern infrastructure that you can deploy to your own environment. This guide covers what you'll need.

## Stack

Orca Memory consists of:

| Component  | Technology       | Purpose                      |
| ---------- | ---------------- | ---------------------------- |
| Dashboard  | React + Vite     | Web interface                |
| Backend    | Convex           | Database, auth, real-time    |
| Embeddings | Python + FastAPI | Vector embeddings for search |
| Auth       | Better Auth      | Authentication               |

## Prerequisites

### Convex Account

Orca Memory uses [Convex](https://convex.dev) as its backend. You'll need:

1. A Convex account (free tier available)
2. A new Convex project for Orca Memory

Convex handles:

* Database storage
* Real-time subscriptions
* Serverless functions
* File storage

### Embedding Service

For semantic search, you need an embedding service. Options:

* **Self-hosted** - Run our Python embeddings service
* **OpenAI** - Use OpenAI's embedding API
* **Other providers** - Any service that returns vector embeddings

### Domain & SSL

For production deployment:

* A domain name
* SSL certificate (use Let's Encrypt or your provider)

## Hardware Requirements

### Minimum (Development)

* 1 CPU core
* 1 GB RAM
* 10 GB storage

### Recommended (Production)

* 2+ CPU cores
* 4 GB RAM
* 50 GB SSD storage

The embeddings service is the most resource-intensive component. Consider running it on a GPU-enabled instance for better performance.

## Next Steps

<CardGroup cols={2}>
  <Card title="Docker Deployment" icon="docker" href="/self-hosting/docker">
    Deploy using Docker Compose
  </Card>

  <Card title="Environment Variables" icon="key" href="/self-hosting/environment">
    Configure your deployment
  </Card>
</CardGroup>