Update README.md

#10
by oql - opened
No description provided.

Indeed. It would be super helpful to have suggested RAM and VRAM for each model, along with expected performance metrics such as how many seconds for a response to a basic query for a given system. Not everyone can afford chonky GPUs that cost a fortune. For instance, I've been running your qwen3:30b model on my laptop with 32GB RAM and a 4GB VRAM integrated 3050 Ti Mobile for about a year just fine, but it can take up to 2 minutes for a response, which I'm completely fine with.

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment