Why is Gemini 4 Carbon taking so long?

(kernel.dev)

54 points | by deenadz 3 hours ago

12 comments

  • tristanj 2 hours ago
    It's pretty obvious that Gemini 4 Carbon is not token efficient. It also has relatively low tokens-per-second compared to other models. I can think of several reasons why this is the case:

    1. Google has huge amounts of cash and compute. Their goal was to train a model that that has the highest quality & frontier performance, not create a model that's efficient to run and serve. They have other models for that, like Hydrogen.

    2. Google created Carbon without caring for inference costs. DeepMind is subsidized by the rest of Google's empire. Selling inference is an insignificant part of Google's revenue. But for Anthropic and OpenAI, selling inference is their entire business model. Anthropic and OpenAI have enormous incentive to sell as much inference as they can in the most efficient manner. Google doesn't have this incentive.

    3. They are one of the few companies that can be inefficient. Out of all the frontier labs, Google DeepMind is probably the least constrained by compute. Chinese labs have around 10 to 100 times less compute, so they must release highly optimized models.

    • fsh 1 hour ago
      4. Google did a worse job than Anthropic and OpenAI.
      • dpe82 1 hour ago
        Well sure, but parent is trying to reason about why.
        • andy_ppp 9 minutes ago
          It’s unlikely to be deliberate tbh, it’s more likely the data, algorithms and the training environments aren’t quite as good as Anthropic, Open AI is a lot closer to Google than people want to admit.
    • zarzavat 1 hour ago
      It's 1. Carbon was announced the day of Google's earnings call. It's a PR stunt.
    • teknologist 1 hour ago
      Also, Google uses their own TPU architecture rather than GPU/Nvidia.
  • horacemorace 2 hours ago
    I was just using this one today for some coding tinkering and it was super slow. But the way it solves problems with a light touch while still driving toward sensible architecture is refreshing after claude’s bulldozer attitude. It was able to integrate repeated creeping scope nicely with what it had already done. Very different feel than sonnet/opus, and even gemini 3.8.
  • NoahZuniga 55 minutes ago
    Google pretty clearly says in their product documentation that carbon is just a version of carbon that thinks way longer.
    • kzrdude 26 minutes ago
      Then that's "test-time scaling" again and Google then showed that they can reach frontier level benchmark results using it. It's pretty cool as a technology actually (the demonstrated capability), even if the model is not so useful when it's slow.
    • lukax 43 minutes ago
      You probably meant a version of Argon?
    • joedwin 47 minutes ago
      Argon right?
  • lp92 2 hours ago
    >Carbon defended against cases the ticket's facts had already excluded, hardening the code against problems nobody had reported.

    That's pretty interesting. Seems like it's over analyzing. Not the worst outcome though to have additional hardening.

    With that said, I found Argon to be better for most coding tasks and Carbon for architecture/deep RCA analysis of complex issues.

  • SubiculumCode 2 hours ago
    That isn't a good look, and either this is unmeasured uncontrolled behavior, or intentional under some longer term plan...but the latter possibility seems improbable.
    • ahsillyme 2 hours ago
      Uhm, I'm not a programmer but why isn't this a good look? I would want my engineers to fix problems that they notice in passing. (Edit: the "unnecessary work" angle is what irked me in the article)
      • horacemorace 1 hour ago
        Good SWEs get a feel for the minimum amount of changes to fix something. Fewer changes usually minimizes the chance for new problems to pop up. Junior SWEs usually try to do too much.
  • Lerc 1 hour ago
    A Theo vid suggests there's a time to first token issue. When doing subagent actions it can compound to be massive from a bunch of internal short queries.

    It really looks like something is broken more than just being inefficient.

    • 96khz 1 hour ago
      are we talking about token prefill here?
      • Lerc 14 minutes ago
        11 seconds worth?
    • teknologist 1 hour ago
      Theo seems to dislike anything that isn't Anthropic Claude. Suspicious, to say the least. On CursorBench, a good measure of real-world coding tasks, Gemini 4 Carbon comes out top: https://cursor.com/evals
      • LiamPowell 52 minutes ago
        Every time I see a Theo tweet he's judging something on the most absurd metrics. Today for example, he was saying Astra 6.2 is bad because when he gave it a single line prompt telling it to create a new level for a game it used a name for the level that he'd seen before. Judging a model on how often it reuses a name is one of the most ridiculous benchmarks I've ever seen.

        As far as I can tell, all his judgements are based on similarly absurd metrics, I don't know why people treat his preferences as valuable.

  • nandanadileep29 44 minutes ago
    yeah i tried using it its not efficient at all. just burns up the token and not get the work done.
  • jrflo 2 hours ago
    Is this due to Google wanting to catch up on benchmarks at all costs, and trains their models to be overly meticulous to do so?
    • murderfs 18 minutes ago
      I'd guess that it's because Google eats its own dogfood, and if they didn't train the models to be overly meticulous, the existing codebase would implode under a tsunami of slop.
  • user_7832 2 hours ago
    I am increasingly more convinced every day that every LLM company is just making their latest model to be the best at (coding) benchmarks, at the cost of pretty much everything else.

    I use LLMs for non code purposes (probably not very common on HN) and Sonnet 3.5/3.6 still ranks as probably one of the best models I've used.

    Can today's models put out tech demos and call more tools that make it way more impressive? Sure. But it's really hard to see any actual significant gains over the last few months, they're so mild (again, non coding).

    • SyneRyder 45 minutes ago
      Are you able to give hints to some of those use cases? Recent Opus and Fable seem to have made big leaps in art and music, and I don't think all of that is the longer context window since Sonnet 3.x. But those are both things that can be created with code and mathematical thinking as well. I definitely wouldn't want to use recent models for writing though (not that I ever wanted them to write for me anyway).
    • auyez 1 hour ago
      At this point it might make sense to put all their resources into improving coding and math skills, because it might lead them to a state where they keep accelerating their capabilities. Even in the worst scenario where LLM approach would hit some wall, they would have technology to quickly explore all other approaches.
    • wellcolormesurp 1 hour ago
      They are cracking software and math, not finance, law or medicine or whatever it is you do.

      Those two domains have leverage, the others do not. It makes perfect sense. If you and I had the goals they have we would start with code and math as well. The other dominos will fall automatically once you got the engineering automated. It makes no sense to stop focusing on that until you either run out of money or achieve your goal.

    • incognition 2 hours ago
      Oddly flash 3.8 for me is good. they really pushed carbon to some other extreme
  • dario_amodei 1 hour ago
    how much?
  • sodowork 1 hour ago
    [flagged]
  • georgestones 1 hour ago
    [dead]