An Old Laptop’s Only M.2 Slot Was Used To Hook Up An RX 7900 XT, Achieving 60 Tokens/s In Qwen3.6 27B & 100K Context Window; External Drive Was Used For The OS

A Lenovo laptop owner connected an AMD Radeon RX 7900 XT through the machine's only M.2 slot using a PCIe adapter.

After booting Windows from an external drive, they ran Qwen3.6 27B locally with llama.cpp at roughly 55–60 tokens per second and a 100K context window.

The setup shows that older laptops can be repurposed for demanding AI inference through unconventional external-GPU configurations.

System RAM capacity and limited expansion options remain major constraints.