Magnitude
July 29, 2026 · View on GitHub
An open source coding agent and inference engine built for local models.
Get started
npm install -g @magnitudedev/cli
cd your-project
magnitude
Magnitude supports macOS and Linux. Windows is supported through WSL.
How it works
Automatic model setup
Magnitude profiles your hardware and recommends the best models your machine can run. Choose Balanced, Best Quality, Fastest, or Lightweight, and Magnitude handles the download and configuration.
An inference engine built for agent work
Magnitude includes a custom inference engine written in Rust on top of llama.cpp. It offers verified model configurations, calculates memory requirements before loading, and tunes acceleration, placement, and batching for your hardware. Parallel agents retain full context windows, model switching preserves consistent tool use, and new requests remain responsive while other work is running.
A coding agent built around local inference
Magnitude can inspect and edit files, run commands, work with images, manage long-running tasks, and delegate work to parallel agents. Because local inference is built in, it also manages model loading and switching and surfaces native prefill, cache reuse, and generation performance directly in the agent UI.
Magnitude Cloud
Magnitude Cloud gives you access to open models that are too large to run on your hardware. It costs $10 for the first month, then $20 per month.
- GLM 5.2, Kimi K3, DeepSeek V4, and more
- Generous usage limits for long coding sessions
- Zero data retention on your prompts and code
- Exa web search for external research
Learn more
License
Magnitude is licensed under the Apache License 2.0.