LLMs on DIY hardware

The ULTIMATE Raspberry Pi 5 NAS (opens in a new tab)

I Built a CoPilot+ AI PC (without Windows) (opens in a new tab)

Run Local LLMs on Hardware from $50 to $50,000 - We Test and Compare! (opens in a new tab)

RPI Llama 3.1 too slow. CPU only mode.

Ollama in a RASPI | Running a Large Language Model in a Raspberry Pi (opens in a new tab)

Local LLM on Raspberry Pi (opens in a new tab)

Phi-2 RAM usage: 1.2GB Speed: 5.5tps