Ask HN: Open Models are 9 months behind SOTA, how far behind are Local Models?

11 points by myk-e 126 days ago | 12 comments

A local model is an open model you run locally, so I'm not entirely sure the distinction in the question makes sense.

That said, if you're talking about models you can actually use on a single regular computer that costs less than a new home, the current crop of open models are very capable but also have noticeable limitations.

Small models will always have limitations in terms of capability and especially knowledge. Improved training data and training regiment can squeeze out more from the same number of weights, but there is a limit.

So with that in mind, I think such a question only makes sense when talking about specific tasks, like creative writing, data extraction from text, answering knowledge questions, refactoring code, writing greenfield code, etc.

In some of these areas the smaller open models are very good and not that far behind. In other areas they are lagging much more, due to their inherent limitations.

myk-e 125 days ago | |

Yes, I meant ordinary hardware which you find at home, like a current MacBook Air or equivalent Windows desktop. There must be a time frame when early SOTA LLMs were at a level that compares to open models that can run on ordinary hardware. But it's more like years rather than months. My rough guess would be 2-3 years. Which still would be amazing if we could get OPUS 4.5 quality within 2-3 years on an ordinary computer.

karmakaze 125 days ago | | |

I don't know if you'd consider this ordinary, but a single Mac Studio M5 Ultra 512GB (or even 256GB) V/RAM seems pretty sweet.

hasperdi 125 days ago |

Well, it depends on the hardware you have. If you have a hardware locally that can run best open models, then your local models are as capable as the open models.

That said, open models are not far behind SOTA, less than 9 months gap.

If what you're asking about those models that you can run on retail GPUs, then they're a couple years behind. They're "hobby" grade.

myk-e 125 days ago | |

Thanks, yes, I meant even ordinary retail PCs, not specialized GPUs. At some point in time in history, SOTA closed models were at a level that compares to todays open models that can run on ordinary hardware.

hasperdi 125 days ago | | |

Retail PCs will probably never catch up to even the open‑weight models (the full, non‑quantized versions). Unless there’s a breakthrough, they just don’t have enough parameters to hold all the information we expect SOTA models to contain.

That’s the conventional view. I think there’s another angle: train a local model to act as an information agent. It could “realize” that, yeah, it’s a small model with limited knowledge, but it knows how to fetch the right data. Then you hook it up to a database and let it do the heavy lifting.

segmondy 124 days ago |

Local models are not behind. There are many specialized local models on huggingface that can do things that none of the closed/commercial models can do. The only way to get that edge is to run locally. When I say many, I mean in the thousands.

myk-e 124 days ago | |

Yes, fair point. I was trying to use the same comparison we are currently having between closed weights and open weights and their time gap. If there might be a similar time gap to what is possible with ordinary equipment.

softwaredoug 125 days ago |

A local model is a smaller open model, so I’d expect it to be 9 months behind a small (ie nano) closed model as a base assumption

myk-e 125 days ago | |

Yes, a small open model that can run on today's hardware and that compared to a historic SOTA closed model with all in. What time difference do we think?