Post
  1. Portrait of Onur Solmaz

    The ideal model could run locally

    @onusoz · /2026/08/14 · View on
    Sam Altman repeated many times that the ideal model for him is a model that does not have a lot of world knowledge, is small by comparison and very cheap to run, and can reconstruct knowledge via reasoning as need arises A paradox with that statement is that, if it were possible to construct such a model, then it also could be small enough to run locally. So why should people pay premium for your proprietary model, if they could have an open weight one running locally for free (*or at the price of electricity) (***there is actually a valid answer to this question, and that is speed, which OpenAI indeed started to focus on recently) DS4 Flash and Qwen 27B line-up lately has been converging to this Sam Altman's platonic ideal of a model—the Altmanic model, if you will Not rented proprietarily, but distributed and hacked on truly freely
    @LLMJunky ·
    AS WE SAID ~Opus 4.6 at home. 42.2 on DeepSWE puts this tiny model somewhere near GLM 5.2 on coding related tasks. This is nuts! Obviously there are benefits to larger models that you will notice compared a 27B model. Some of you are so pedantic. You completely miss the point. The point is that for many tasks, especially tasks that are useful for doing real work, you can get REALLY close. At home. On your own hardware. That is amazing!!! Will you use this for the most difficult engineering tasks? No of course not. But guess what? You're not using Opus 4.6 either! For any task that you would delegate to such a model, this model will also do great. I received a ton of comments about being skeptical about how good this would be, but this was the only logical result. Qwen 3.6 27B was still among the best small models, any improvement would have made this SOTA for its size. So this should be no surprise, it's SOTA for its size. :) Sorry for the victory lap, but i'm just excited. This is great news for everyone.