Run This Ai
EN DE

Moltis

A secure persistent personal AI agent server in Rust — one binary, sandboxed execution, multi-provider LLMs, memory, and MCP tools

★ 2,781 GitHub MIT ai-agentrustvoice-assistantmcpself-hostedsandboxpersonal-ai AI Agents

Overview

Moltis is a secure, persistent personal AI agent server built entirely in Rust. Unlike complex agent frameworks that require Node.js, Python runtimes, or heavy dependencies, Moltis runs as a single self-contained binary on any hardware — from a Raspberry Pi to a Mac Mini to a cloud server. It features sandboxed execution where your API keys and data never leave your machine, multi-provider LLM support (OpenAI, Anthropic, Ollama, and more), built-in voice input/output, persistent memory with SQLite+FTS+vector search, cross-session recall, automatic edit checkpoints, scheduling, and deep integration with Telegram, WhatsApp, Signal, Discord, Microsoft Teams, Slack, and Matrix. Moltis also supports MCP (Model Context Protocol) via stdio and HTTP/SSE, allowing you to extend its capabilities with thousands of MCP servers. The entire Rust workspace is ~270K lines across 59 modular crates, designed for auditability and performance.

Requirements

Min vCPU
1
Min RAM
512 MB
Min Disk
10 GB
Rec vCPU
2
Rec RAM
1024 MB
Rec Disk
20 GB

Recommended VPS

Hostinger · KVM 2

2 vCPU · 8192 MB · 100 GB

$6.99
View plan

Hostinger · KVM 2

2 vCPU · 8192 MB · 100 GB

$6.99
View plan

Hostinger · KVM 4

4 vCPU · 16384 MB · 200 GB

$11.99
View plan

Affiliate disclosure

Docker Compose

# Generated by Run This Ai — docker-compose.yml
services:
  moltis:
    image: ghcr.io/moltis-org/moltis:latest
    restart: unless-stopped
    ports:
      - 8080:8080
    volumes:
      - ./data/moltis:/data

Best VPS for Moltis →

Related tools

Guides & articles