What resources to train from scratch

[reynoldscem]

Hi,

Could you please provide some numbers around which GPUs, how many of them, and how many hours roughly were required to train these models?

[ZexinHe]

Hi,

Plz refer to this issue https://github.com/3DTopia/OpenLRM/issues/2#issuecomment-1882590904. There isn't much differences for V1.1 models.

Plz also note that you can also use less resources, e.g. 8 A100 GPUs, by configuring the gradient accumulation steps in the config file.