RiQuY is asking if using the flags --model and --chat-model simultaneously will load both models in GPU at the same time, causing the VRAM costs for each model to add up.