I tried the 27b-iat model on a 4090m with 16gb vram with mostly default args via llama.cpp and it didn't fit - used up the vram and tried to use about 2gb of sy

I tried the 27b-iat model on a 4090m with 16gb vram with mostly default args via llama.cpp and it didn't fit - used up the vram and tried to use about 2gb of system ram: performance in this setup was < 5 tps.