Token Generation

topic on 2 shows · 2 statements across 2 episodes

David Senra the MAD Podcast

2 statements about Token Generation, every show

Ross: Splitting LLM pre-fill and generation across different chips is a mistake
“What we realized was, and this is what most people get wrong when they're trying to do this themselves. They'll, they'll take the reading of tokens or what's called pre-fill, and they'll do that on one piece of hardware, and then they'll put the generation of …”
Jonathan Ross Jul 5, 2026 ▶ 27:23 Why Asking the Right Questions Is the Most Important Skill in the AI Age
MAD Insight
LLM prompt processing is compute-bound; next-token generation is memory-bound
“Prompt processing is bottlenecked by computation, and generating next, predicting next token is bottlenecked by memory bandwidth.”
Lin Qiao Mar 27, 2025 ▶ 30:42 Why This Ex-Meta Leader is Rethinking AI Infrastructure | Lin Qiao, CEO, Fireworks AI

← every entity, every show

Made with StarZero

Turn any episode into a week of clips.

This entire site, thousands of episodes across every show transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.