Autonomous Computer: Open-Source Build Guides for Local AI Workstations

384GB of VRAM. On your desk. Not in a datacenter. Autonomous, the company behind SmartDesk, just open sourced complete build guides for a personal AI computer that runs open models on hardware no one can switch off. It's called Autonomous Computer. Three configurations, one philosophy: a model you rent can be cut off overnight. A model running in your own house can't. → Home build: 2x RTX 5090 workstation → Team build: 4x RTX PRO 6000 Blackwell, 384GB VRAM → On-prem business build: 8x RTX 4090/5090, up to 256GB VRAM Every config ships with the full bill of materials, 3D printable and CNC housing files, wiring diagrams, BIOS tuning, and assembly photos. Software side covers OS setup, NVIDIA drivers, and serving open models through Ollama, vLLM, and llama.cpp. Fork it. Change it. Build it. Sell it. MIT License. 100% Opensource. Repo: https://lnkd.in/gA6z_E4k

This post is based on a real open-source project and accurately describes many of its features. However, a few statements are promotional or simplify the trade-offs of running local AI hardware.

Confirmed Accurate

• Autonomous has open-sourced build guides for Autonomous Computer, a DIY personal AI workstation project.
• The project includes multiple hardware configurations for individuals, teams, and on-premises deployments.
• The build guides provide detailed documentation, including bills of materials, assembly instructions, wiring diagrams, and enclosure files.
• The software setup covers installing the operating system, NVIDIA drivers, and serving models with tools such as Ollama, vLLM, and llama.cpp.
• The project is released under the MIT License.

Not Fully Accurate / Promotional

• The 384GB of VRAM claim applies only to the high-end team configuration, not every build.
• The statement that "a model running in your own house can't be switched off" is a philosophical opinion about self-hosting rather than a technical fact. Local deployments still depend on hardware, electricity, and software maintenance.
• The phrase "Sell it" refers to the permissive MIT license, but users must still comply with the licenses of the hardware, software, and AI models they use.
• Performance and the size of models that can run locally depend on the chosen hardware configuration and model requirements.