Qwen3.8 27B Q4 on my low-end Laptop, and it became my main Model usage for coding

my laptop is very old, like 20G of ram 4G Vram 256G NVMe Gen3 From samsung i7 8th gen on these specs i got around 0.67 - 0.77 token per second yes, extremely slow, most of you won't even like it, or use it but free APIs