Strata says it runs a 125-billion-parameter AI model on gaming PCs

The open-source Strata project says it runs Qwen3.8-Flash-Next, a 125-billion-parameter AI model, locally on Windows or Linux PCs with NVIDIA or AMD graphics cards containing at least 12 GB of memory. It supports chat, coding and connections to applications through compatible APIs. Strata distributes work across the graphics card and processor, keeping frequently used model specialists on the card and the full set in system memory. A smaller helper model guesses output that the larger model checks. The standard setup downloads about 70 GB and loads 35-55 GB into RAM; some larger versions rely on SSD reads and run more slowly.