10-17-2026, 11:31 AM
I'd start with the simpler stack.
2x3090 • llama.cpp • too much VRAM is still not enough
|
context size vs speed in GGUF: setup advice
|
|
« Next Oldest | Next Newest »
|