Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Jev-rs: a Rust crate to turn a LLM to serve jev(github.com/yijunyu)
    1comments
  2. The Moon in the Terminal(isene.org)
    discuss
  3. Pyrowave video codec now in beta(steamcommunity.com)
    discuss
  4. Show HN: Red-team your AI agent(botgauge.com)
    discuss
  5. OpenTelemetry SDK Overhead(coroot.com)
    discuss
  6. Just the Links, Please(brettterpstra.com)
    discuss
  7. Blink: An Embeddable Jev Engine(github.com/sqliteai)
    discuss
  8. Show HN: BuildABrowser (Java) – Passes Acid2, Has CSS Grid(buildabrowser.net)
    discuss
  9. Apple Links Landmarks on Its Maps App to Hidden Histories Podcast Episodes(engadget.com)
    discuss
  10. The body temperature of Tyrannosaurus rex(science.org)
    discuss
  11. Dark matter detection: Event LZ230616(wikipedia.org)
    discuss
  12. Jev-serve: Run latest qwen3.8, other frozen LLM models, mlx-supported(github.com/rreinold)
    discuss
  13. Top free SaaS listing directories(goodsaas.xyz)
    discuss
  14. Dyson Won't Say What's Wrong with Its CameraJet Toothbrush(wired.com)
    discuss
  15. Early-2000s investor decks were their own art form(slidebook.io)
    discuss
  16. GPT-6 Astra is the best vision model(roboflow.com)
    discuss
  17. What if we created a local AI tool to measure your security?(github.com/just-not-google)
    discuss
  18. Explaining ShapeshiftUI – Jev picks the UI, TypeScript does the math(qala.lol)
    discuss
  19. How to Optimise Hiring for Software, Platform and SRE Roles(noxidjobs.co.uk)
    discuss
  20. WTF Is Baoism? (In English, Hindi, Japanese, Chinese, Thai)(xcrol.com)
    discuss
  21. We should share childhood stories more(devz.cl)
    discuss
  22. Does Consciousness Break the Laws of Physics?(skeptic.com)
    discuss
  23. New Nox night train from Munich to Hamburg(back-on-track.eu)
    discuss
  24. Should we leave encouragement notes for AIs everywhere?
    discuss
  25. Samsung's AI fridges shut down after update, causes outrage(notebookcheck.net)
    discuss
  26. Meta Tests Muse AI Agent Calls That Are Made by Humans in a Call Center(404media.co)
    discuss
  27. Tiny OLED pixels can be manufactured like microchips(ethz.ch)
    discuss
  28. Two Git ignore files nobody told me about(dinculescu.dev)
    2comments
  29. Universal Agent HP – 10-layer autonomous AI agent OS (free or local)(github.com/kapitan00000978-sketch)
    discuss
  30. Show HN: An Android app that goes below the phone's minimum volume(granularvolume.com)
    discuss

Ask HN: Do you think we'll ever have local models of Fable level?

2 pointsby 1h ago
2 comments
I'm not talking only about Fable level models, since we are obviously already close with stuff like Qwen.

But I'm also wondering about being able to run them on consumer-end hardware.

I remember using a local model 2-3 years ago and had to wait around 2-3 minutes for a basic answer to be printed. Now I'm running a "thinking" Qwen on a 16GB GPU and I'm able to do anything I'd do with Opus a couple months ago, at nearly the same speed. But that does use my entire VRAM and most of the RAM I have. No way I can also run a game or something else on the side.

But like how we went from bulky PCs to smartphones 1000x faster, and at the rate local models already improved, do you think we'll ever be able to have the same kind of models running locally, on affordable hardware, on our phones or maybe our fridges?

Not saying we should use them on anything, that will be a question for later, but strictly thinking about capabilities.

37m agoHN ↗

but strictly thinking about capabilities.

Yes, trivially so.

If the investment currently driving new models ceases for whatever reason, there's plenty of ways to burn weights into read-only hardware that's much more energy efficient and compact, we don't even need new lithography nodes for smaller transistors.

Only a few are attempting this today, because the model update cycle is so fast compared to mass production of hardware, that any wights you burn into hardware are likely obsolete before you can ship them.

If you want to get sci-fi about it: what happens when we can take a biopsy from someone to get some stem cells, use it to tissue culture a brain organoid, use a nano-electrode "neural lace" to send the right electrical impulses to exploit biological processes so as to wire up the synapses to encode the same patterns as some AI model (probably not a Transformer architecture, those are situationally good for computers and not likely the best for biology), then implant it in the head of whoever you got the tissue sample from?

With currently reported model sizes, at 1 param ~= 1 synapse, this would be around 1% of your normal brain volume.

Everyone could skip further education while also getting to at least "pass the exams" level knowledge in… how many subjects can LLMs already pass exams in? Does anyone even check this since GPT-4?

https://en.wikipedia.org/wiki/Cerebral_organoid

https://en.wikipedia.org/wiki/Long-term_potentiation

28m agoHN ↗

Nothing in the laws of physics prevents it. Heavy regulation might.