NanoGPT Speedrun Frontier

(primeintellect.ai)

44 points | by stared 3 hours ago

8 comments

  • vibe42 1 hour ago
    "Almost every model finds the same winning ideas. What separates the best traces is what an experiment leaves behind. They preserve weak signals long enough to validate them, but they also have a better understanding of the results."

    Curious if a harness that helped preserve signals in some history log would change the outcome.

    Also curious if different goal prompts would have changed the outcome. Not a bunch of prompt engineering; small diffs like "consider novel solutions, keep track of weak signals".

    IMO they allocated quite a bit of GPU time to the same goal prompt.

  • ninjahawk1 1 hour ago
    I might’ve missed it, but why was Fable 5 tested on high while Opus 5 was tested on max? Seems like quite a few of them aren’t on the same effort setting as well. Although effort doesn’t really matter anymore since they can change it dynamically, seems like that might be viewed as an experimental error to some.
  • c0rruptbytes 21 minutes ago
    auto research the new cool kid on the block - look at https://mlx.fast
  • totetsu 1 hour ago
    “We ran 153 autonomous runs across 18 frontier models on the nanoGPT optimizer speedrun.”

    Uh.. okay.. but whats a run… read blog

    “We want to measure how well frontier models can conduct research….””we ran 153 autonomous runs on the nanoGPT optimizer speedrun across”

    Okay but what is a optimiser run and what connection does it have to being good at research?

    “For comparison, Anthropic's internal automated AI R&D evaluation optimizes a model on a CPU node,”

    So I should go look what Anthropic was doing to understand?

    Why not just explain what it means in their blog..

    • totetsu 52 minutes ago
      So maybe this is a simple way to put it..

      They gave 18 frontier models the task of “researching” how to improve a lab-rat nano model’s training. Stopping when it met a quality goal of a target loss rate. During each autonomous research session, the AI repeatedly tried changes, tested them, and used the results to decide what to try next. They repeated the whole research session many times with different seeds to average out variance.

    • derac 1 hour ago
      I think that's explained here:

      https://www.primeintellect.ai/blog/measuring-autonomous-rese...

      Basically they do 8 runs trying to optimize to under 3.28 loss in the fewest training steps possible under time/token constraint. I dunno why 18 * 8 != 153 (it's 144)

  • skybrian 1 hour ago
    Neat!

    The graphs show the "best validated result" for each model. I wonder how much variation there is between runs for a model?

  • logicallee 19 minutes ago
    I'm highly interested in their Grok 4.6 run which is currently running. I think it is a very good fit for doing really well on this benchmark. Will have to check this page again in a couple of days.

    I wrote a quick review of Grok 4.6 here: https://taonexus.com/publicfiles/aug2026/grok-4-6-review/

  • moffers 19 minutes ago
    “We put liquid bandage on a bee, and it died”
  • ninjahawk1 1 hour ago
    I misread the graph and genuinely thought you put NanoGPT where Fable is.

    Lol.

    • kelseyfrog 1 hour ago
      I misread the title and thought it would be about the (for lack of a better term) NanoGPT speedrun[1]. Which, previous to the article, was meant to be the world speed records for Andrej Karpathy's GPT-2 (small) reproduction.

      1. https://github.com/KellerJordan/modded-nanogpt#world-record-...

      • cookiengineer 1 hour ago
        You're not the only one. I thought so too.

        I just ran it the last couple days extensively to verify my data training pipeline I'm building for my gonano SIMD port.

        Given that the speed records and the runs are sponsored by the same company I was confused a bit.