A

Arogya • 3.19K Points
Extraordinary

Q. Why is GPU VRAM important when running LLaMA?

  • (A) Stores keyboard input
  • (B) Stores model weights during inference
  • (C) Controls internet access
  • (D) Controls operating system booting
  • Correct Answer - Option(B)
  • Views: 3
  • Filed under category Llama
  • Hashtags:
Explanation by: Arogya
The model parameters must fit in VRAM for fast inference.

You must be Logged in to update hint/solution

Discusssion

Login to discuss.

Be the first to start discuss.