Ask a question below.
Published answers will appear here.
Chamath on all important “prefill” and “decode.” in AI compute. Prefill is compute-bound; massive parallel GPUs win, so Nvidia dominates as context grows. Decode is memory-bandwidth bound as each next token depends on scanning what’s already generated https://x.com/theallinpod/status/2007917878519931285/video/1
Chamath on all important “prefill” and “decode.” in AI compute. Prefill is compute-bound; massive parallel GPUs win, so Nvidia dominates as context grows. Decode is memory-bandwidth bound as each next token depends on scanning what’s already generated https://x.com/theallinpod/status/2007917878519931285/video/1
Ask a question below.
Published answers will appear here.