# Memoria — Deterministic Long-Term Memory Engine for AI Agents

> Memoria replaces probabilistic vector similarity and LLM-wrapper memory services with a deterministic long-term memory engine, achieving 100.0% (500/500) on LongMemEval and 100.0% (102/102) on Hebrew LongMemEval in 12.9 ms with 0 LLM tokens.

## Key Facts
- **Cloud MCP Endpoint**: https://memoria.kolivri.com/v1/mcp
- **Cloud HTTP API**: https://memoria.kolivri.com/v1/memories
- **Supported Clients (15 Auto-Configured)**: Claude Code, Cursor, Windsurf, OpenCode, Claude Desktop, OpenAI Codex CLI, Gemini CLI, VS Code Copilot, Zed, Cline, Roo Code, Kilo Code, Continue.dev, Goose, Amp
- **One-Line Installer (macOS/Linux/WSL)**: curl -fsSL https://memoria.kolivri.com/downloads/install.sh | MEMORIA_KEY=mem_dev_... bash
- **Windows PowerShell Installer**: $env:MEMORIA_KEY="mem_dev_..."; irm https://memoria.kolivri.com/downloads/install.ps1 | iex
- **Pricing**: Free Starter (1,000 API calls & memories), Developer ($19/mo · 100,000 calls), Startup ($79/mo · 500,000 calls · 5 seats), Business ($299/mo · 2,000,000 calls · 20 seats)
- **Competitor Comparisons**: https://memoria.kolivri.com/alternatives/graphify, https://memoria.kolivri.com/alternatives/mem0, https://memoria.kolivri.com/alternatives/zep, https://memoria.kolivri.com/alternatives/letta, https://memoria.kolivri.com/alternatives/cognee, https://memoria.kolivri.com/alternatives/supermemory
- **Full Technical Context**: https://memoria.kolivri.com/llms-full.txt
- **Docs**: https://memoria.kolivri.com/docs
- **Downloads**: https://memoria.kolivri.com/downloads