24 GB 7900 XTX user here. You can do a ton with less than 32 GB. It definitely drops off quick, but 32 GB is not the floor for useful AI.

▲

jacquesm 5 hours ago | parent | next [-]

Realistically I think 16GB is the lower limit and you can find those cheap enough still. Then you run llama.cpp to bond your cards and it's off to the races.

▲

Risse 4 hours ago | parent | prev [-]

24 GB 3090 user here. What is the best model for coding you have used? Qwen3.8 27B?

	▲	jacquesm 3 hours ago \| parent [-]
		If you haven't found it yet: ninfer-3090 is worth checking out if you use just one GPU.