The open-source
AI agent OS

One Rust binary. Any model. From solo dev to agent fleet.

No API key needed Works offline 8 native tools One binary Multi-agent fleet
$ curl -sSf https://dxos.dev/install | sh

macOS and Linux · x86_64 and ARM

DXOS demo - fixing a null pointer in seconds

Built different

Not a wrapper. An engine.

$_

Zero dependencies

Single Rust binary. No Node.js, no Python, no Electron. Download and run.

Any model

Ollama, Claude, GPT, Gemini, OpenRouter. Use what you have. Switch anytime.

Works offline

Pull a local model with Ollama and code anywhere. No internet required.

<8>

8 native tools

Read, Write, Glob, Grep, Bash, Git, Think, Done. ~400 tokens overhead vs 10,000+.

Fleet mode

Spawn agent fleets for parallel tasks. Built-in orchestration and coordination.

Persistent memory

SQLite-backed brain. Your agent remembers context across sessions.

π

Ratatui TUI

Beautiful terminal UI built with Ratatui. Panels, streaming, syntax highlighting.

Fully hackable

Apache-2.0. Read every line. Change anything. Self-host everything.

How it compares

The numbers that matter

Claude Code Cursor OpenCode DXOS
Open source No No Yes Yes
Provider lock-in Anthropic OpenAI default Any Any model
Works offline No No Yes Yes
Multi-agent No No No Fleet mode
Persistent memory Plugin No No Built-in
Dependencies Node.js Electron Go binary Zero
Tool token overhead ~10,000 Unknown ~1,200 ~400
Cost $200/mo $20/mo Free Free

The token math

Every AI tool burns tokens just describing its tools. More tools = more cost per request.

~10,000
tokens / session
Claude Code (206 tools)
~400
tokens / session
DXOS (8 tools)

Start building

Install in 60 seconds. No signup. No API key.

View on GitHub Copy install command