How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip

(spectrum.ieee.org)

34 points | by maxall4 1 hour ago

5 comments

  • pama 19 minutes ago
    Having worked with people doing bringup of specialized chips, I am awed at how the world has changed.

    > When the first chips came back from the foundry in May, the team pointed its internal AI models at designing software to run benchmarks such as SemiAnalysis’s InferenceX. On DeepSeek’s multi-head latent attention kernel benchmark, performance climbed from 0.31 percent of the theoretical ceiling (set by the chip’s compute and memory bandwidth) to 88.94 percent in roughly 40 hours. Ho says this result is repeatable, so the time between when foundries deliver the first chips and when production ramps up can be reduced. “All our schedule assumptions are going to be based on the fact we have this capability now,” he says.

  • geraneum 2 minutes ago
    Whatever happened with the Apple lawsuit?
    • Legend2440 0 minutes ago
      Still happening? Lawsuits often take years to argue through the legal system, it may be half a decade before it resolves.
  • karim79 1 hour ago
    I grow Jalapeños. This conflation of AI and actual chili peppers irks me.
    • amelius 32 minutes ago
      Guess how electrical engineers feel about the term "transformers".
      • frangonf 13 minutes ago
        As a former EE, attention was all I needed to not get zapped.
      • karim79 29 minutes ago
        This is an excellent comment. I'm still laughing.
    • Lalabadie 1 hour ago
      I do generative art (no relation to AI prompting). I feel your frustration.
      • fragmede 13 minutes ago
        Cryptographers also got the same raw deal
      • TomGarden 29 minutes ago
        Oh my!
    • seanmcdirmid 11 minutes ago
      Jalapeño also used to be a Java VM written in Java at IBM.
    • asveikau 1 hour ago
      Just think of how the people of Xalapa, Mexico feel. They should send them a royalty check.
    • glitchc 34 minutes ago
      Feeling the burn?
    • honeycrispy 33 minutes ago
      I'm annoyed that the meaning of the word "Agent" has been obliterated.

      Like, why couldn't they invent a new word and not hijack an existing word?

      • karim79 20 minutes ago
        Call it GPTChippomatic or something. Please leave my peppers alone.
      • Razengan 24 minutes ago
        Did you not watch the Matrix documentary?
    • Razengan 1 hour ago
      > irks me

      It's jalapeño grill would you say?

      • karim79 58 minutes ago
        Not sure what you're talking about. But I'll tell you, home grown Jalapeño peppers, fermented with 3% salt is the stuff of dreams.
        • wiml 39 minutes ago
          "It's all up in yo' grill, would you say?"
          • Razengan 29 minutes ago
            You know what really grinds my gears? Friction.
  • amelius 1 hour ago
    At some point people will use an LLM to design an Apple M series competitor.
    • Lramseyer 29 minutes ago
      Production grade CPU design is more than just the RTL (the source code.) To achieve the performance numbers that these companies get, you have to do a ton of optimization in your physical design to achieve the power/performance/area (PPA) metrics that make these products competitive. LLMs are not suitable for that kind of work.

      There are people working on PPA optimization and trying to shake up how things are done, just not with LLMs.

      • btown 12 minutes ago
        Something that I think is fascinating, though, is that labs are no longer beholden to the limitations of commercial design software. Want to replace your simulator and optimizer with a fully custom verifiable stack of Lean proofs of optimality and correctness? Just throw your unlimited token budget at it.
    • bhouston 56 minutes ago
      It is probably doable right not to push a risc-v design into that performance space.
    • bigyabai 1 hour ago
      They won't, because they'd need an ARM architecture license.
      • amelius 1 hour ago
        Why, the LLM can make up its own architecture.

        The value lies in the design space exploration, which is what an LLM can easily do.

        https://en.wikipedia.org/wiki/Design_space_exploration

      • wmf 54 minutes ago
        Arm sells architecture licenses to anybody these days.
      • cmrdporcupine 11 minutes ago
        Or they'll just build a competitor in RISC-V instead and that's fine.

        Except the problem is not restricted to the actual ISA or its HDL implementation, etc.

        It's even just getting space / time in a fab at that advanced of a process node.

      • pixl97 1 hour ago
        I mean you can design anything without a license. Selling it is where the problems come up. Even then there are likely places in China that would still make it for you.
  • cute_boi 42 minutes ago
    openai should figure out how to make lithography machine, so ASML don't have monopoly on it.
    • TomGarden 21 minutes ago
      The Chinese have been working on EUV for a while
    • bigyabai 31 minutes ago
      "Reverse engineer this DARPA project, make no mistakes"