Free Open-Source Artificial Intelligence @lemmy.world

tinwhiskers @lemmy.world

2y ago

New technique to run 70B LLM Inference on a single 4GB GPU