‹ BackHN Continuity

Thread

How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip

206 points · 139 comments · maxall4

  1. ramshanker · · focus · HN ↗
    So when can we start getting cheap chips? RAM anyone please!
    1. faitswulff · · focus · HN ↗
      Everyone's still bottlenecked on foundries, not designs.
      1. jcims · · focus · HN ↗
        Isn’t this the whole concept for (braces) terafab? Reduce iteration cycle time.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.