Comment by Depurator Comment by Depurator 5 days ago 0 replies Copy Link View on Hacker News I'm curious what kind of performance the Tensor G5 would have with llama.cpp, compared to a 16gb desktop gpu.