Nvidia with unusually fast coding model on plate-sized chips

(arstechnica.com)

23 points | by Bender 4 days ago ago

27 comments

  • greyskull 2 hours ago

    Missing "OpenAI sidesteps" from the beginning of the title article title

    • sorenbs an hour ago

      Yeah. Completely changes the meaning of the article. I thought Nvidia was now competing with Cerebras. That's not the case...

    • jeron an hour ago

      very excited for cerebras, hopefully nvidia/amd will have less AI sales and bring back more consumer options when they realize they have abandoned/neglected the market that made them who they are

  • nguyentran03 an hour ago

    the hardware diversification story here is more interesting than the speed numbers. OpenAI going from a planned $100B Nvidia deal to "actually we're unsatisfied with your inference speed" within a few months is a pretty dramatic shift. AMD deal, Amazon cloud deal, custom TSMC chip, and now Cerebras. that's not hedging, that's a full migration strategy.

    1,000 tok/s sounds impressive but Cerebras has already done 3,000 tok/s on smaller models. so either Codex-Spark is significantly larger/heavier than gpt-oss-120B, or there's overhead from whatever coding-specific architecture they're using. the article doesn't say which.

    the part I wish they'd covered: does speed actually help code quality, or just help you generate wrong code faster? with coding agents the bottleneck isn't usually token generation, it's the model getting stuck in loops or making bad architectural decisions. faster inference just means you hit those walls sooner.

    • aurareturn an hour ago

      If you are OpenAI, why wouldn’t you naturally want more than one single supplier? Especially at a time where no one can get enough chips.

    • conception an hour ago

      With agent teams I’ve found CC significantly better at catching mistakes on itself before it finishes its task. Having several agents challenging the implementation agents seems to produce better results. If so, faster is better always as you then can run more adversarial/verification tasks before finishing.

    • nerdsniper 33 minutes ago

      I'm 99% sure this 20-hour old user is an LLM posting on HN. Specifically, ChatGPT.

    • irishcoffee an hour ago

      > OpenAI going from a planned $100B Nvidia deal to "actually we're unsatisfied with your inference speed" within a few months is a pretty dramatic shift.

      A different way to read this might be: Nvidia isn't going to agree to that deal, so we now need to save face by dumping them first"

      I imagine sama doesn't like rejection.

  • Havoc an hour ago

    > On Thursday, OpenAI released its first production AI model to run on non-Nvidia hardware,

    They used amd gpus before - MI300X via azure a year plus ago

  • uniclaude 2 hours ago

    Previous discussion on 5.3 codex Spark (sharing as the article doesn’t add tremendous value to it): https://news.ycombinator.com/item?id=46992553

  • gortok 2 hours ago

    Ever since the recent revelation that Ars has used AI-hallucinated quotes in their articles, I have to wonder whether any of these quotes are AI-hallucinated, or if the piece itself is majority or minority AI generated.

    If so, I have to ask: If you aren’t willing to take the time to write your own work, why should I take the time to read your work?

    I didn’t have to worry about this even a week ago.

    • what an hour ago

      >I didn’t have to worry about this even a week ago

      No, you didn’t realize you had to worry about this until a week ago.

      • cyanydeez an hour ago

        Im actually very cpncerned people have yet to realize they dont need to put truth values on internet content.

        Once you default to 'doesnt matter if true' you end up being a lot more even keeled.

  • reliabilityguy 2 hours ago

    I have a question for those who closely follows Cerebras: do they have a future beyond being inference platform based on (an unusual) in-house silicon?

    • dgacmu an hour ago

      My mental model of cerebras is that they have a way of giving you 44GB of SRAM (and then more compute than you'll probably need relative to that). So if you have applications where the memory access patterns would benefit massively from basically having 44GB of L3-ish-speed SRAM, and it's worth $1-3m to get that, then it's a win.

      Honestly not sure what else fits that bill. Maybe some crazy radar applications? The amount of memory is awfully small for traditional HPC.

    • sorenbs an hour ago

      If chip manufacturing advances allow them to eventually run leading edge models at speeds much faster than competition, that seems a really bright future all on its own. Their current chip is reportedly 5nm already, and much too small for the real 5.3-codex: https://www.cerebras.ai/press-release/cerebras-announces-thi...

    • bob1029 an hour ago

      They can also train models using this silicon. They're advertising 24T parameter models on their site right now.

    • wmf 41 minutes ago

      Do they need any future beyond inference? It's going to be a huge market.

      • reliabilityguy 37 minutes ago

        In principle? No. In practice? I think others, eg TPUs and Trainiums of the world will cannibalize a lot of Cerebras’s share. I am not an expert though, that’s why I’m asking opinions of others.

    • adgjlsfhk1 an hour ago

      tldr is possibly. their packaging does offer inherent advantages in giving you maximal compute without external communication, and that seems likely to remain true unless 3d stacking advances a lot further.

  • ElijahLynn an hour ago

    Title is currently: "OpenAI sidesteps Nvidia with unusually fast coding model on plate-sized chips"

  • RobotToaster an hour ago

    One thing I don't get about Cerebras, they say it's wafer scale, but the chips they show are square, I thought wafers were circular?

    • gpm 42 minutes ago

      Their chips aren't actually square, they get an extra 2.9mm in both dimensions by having slightly rounded corners. They are wasting the rest of the circle though yes.

    • wmf 42 minutes ago

      They cut off the sides. It's the largest square you can make from a wafer.

    • cyanydeez an hour ago

      I believe the discs are a product of the manufacturing, having to spin them. The entire disc is not useable, so not really would you call it a wafer. If the entire cjip comes from the wafer, its wafer scale.

  • AndrewKemendo an hour ago

    Mark my words:

    The era of “Personal computing” is over

    Large scale Capital is not gonna make any more investments into microelectronics going forward

    Capital is incentivized to make large data centers and very high speed private Internet, not public Internet, private Internet like starlink

    So the same way in the 1970s it was the main frame era and server side computing, which turned into server side rendering, which then turned into client side rendering which culminated in the era of the private computer in your home and then finally in your pocket

    we’re going back to server side model communication and that’s going to encompass effectively the gateway to all other information which will be increasingly compartmentalized into remote data centers and high-speed access