throwa356262

4 ms·

1. 61.▲
   by throwa356262 1mo ago

   I dont know, warm weather but not too much direct sunlight could actually be a good thing.
2. 62.▲
   by throwa356262 1mo ago

   O3 is supposed to draw 25% less than O1, which according to reviews of their tablets did not throttle. Their next Fold phone (TBA September) will use O3. I guess we will find out very soon.
3. 63.▲
   by throwa356262 1mo ago

   All Canada needs to do to become the center of innovation now is to... become 10°C warmer.
4. 64.▲
   by throwa356262 1mo ago

   Does it test hardware, or hardware plus software that can be different on different platforms?
5. 65.▲
   by throwa356262 1mo ago

   Linus Torvalds has in past used laptops from Apple, Lenovo and Dell and probably many more. His current laptop is an Intel laptop.
6. 66.▲
   by throwa356262 1mo ago

   After this news they might suddenly be.
7. 67.▲
   by throwa356262 1mo ago

   I dont think Geekbench is a good cross-platform benchmark, given what it measures
8. 68.▲
   by throwa356262 2mo ago

   If the hard cores are configured with incorrect cache coherency options you may not be able to correct the issue in fpga fabric. But this is pure speculation, I haven't checked Cyclone V documentation myself.
9. 69.▲
   by throwa356262 2mo ago

   This is about the hard macros in Altera Cyclone V FPGA. I suspect the issue is in how Altera configured the IP. They may simply have disabled some part to save on area.
10. 70.▲
    by throwa356262 2mo ago

    Thanks, now I too want a Lenovo Thinkstation PGX... I think it will be fairly easy to remove refusals from open models. Feels like a lost battle, so why does Alibaba even bother?
11. 71.▲
    by throwa356262 2mo ago

    That is correct, but if we compare the top 10 literally work of the last 20 years to the same period 500 years ago... Do you think the quality has improved? Do you think many people will read Stephen King in 500 years?
12. 72.▲
    by throwa356262 2mo ago

    When you think about it, humans have never been reading as much as we do right now thanks to Internet and digitalisation. Yet the quality has gone down when you compare it to, say, 1600's greatest works.
13. 73.▲
    by throwa356262 2mo ago

    "We were again surprised by the lack of novelty. The models clearly understand the objects they manipulate at a deep level, and yet very few genuinely new ideas emerge..." That was disappointing. I guess 2026 will not
14. 74.▲
    by throwa356262 2mo ago

    Correct link https://xcancel.com/deepseek_ai/status/2090730032574631962
15. 75.▲
    by throwa356262 2mo ago

    This is a fairly small model for coding and agentic work. Training it on images like yours would just make it worse in other areas.
16. 76.▲
    by throwa356262 2mo ago

    For example search and OCR in the gallery and camera app.
17. 77.▲
    by throwa356262 2mo ago

    The NPU in previous AMD generation has unfortunately it's own dedicated RAM that is far too small for an LLM. But even those should comfortably run an SLM of around 100-400M parameters at crazy high speeds and with minimal power usage.
18. 78.▲
    by throwa356262 2mo ago

    I think Samsung has the best amount of AI (once you turn off the Google crap). There are tons of things that work like magic. You don't even notice you are using AI. Even Bixby turned out to be uuseful. Although I have disabled the sho
19. 79.▲
    by throwa356262 2mo ago

    "We also made some smaller UD-1bit quants with UD-IQ1_S being 6.2GB (without MTP) which retain around 72% top-1% accuracy yet being 89% smaller" This is crazy! But has anyone tried these lower quants on real projects?
20. 80.▲
    by throwa356262 2mo ago

    And the 5 years old AMD Ryzen 5 5600H is doing 7M? Am I reading this right? Then I need to try this on Strix Halo
21. 81.▲
    by throwa356262 2mo ago

    At the opposite end of the spectrum we have Sony Mobile, where product names are probably assigned by their worst enemy. You can fully expect them to release a high end phone called Xperia one II 10 as successor to last year's
22. 82.▲
    by throwa356262 2mo ago

    This breaks a few things or at least make things harder. Also, if every app would save their carp in my home folder it soon would be impossible to find anything.
23. 83.▲
    by throwa356262 2mo ago

    I dont think this is benchmaxing. They have simply decided to not train the model in some areas such as world physics
24. 84.▲
    by throwa356262 2mo ago

    Have not read the paper yet, but this not sound like GAN applied to agent training?
25. 85.▲
    by throwa356262 2mo ago

    This seems to be the easiest way to get on HN front page: Come up with an arbitrary test, let a bunch of LLMs work on it. Make some very subjective judgement about the result...
26. 86.▲
    by throwa356262 2mo ago

    4) Elon has access to some Anthropic models because Dario is desperate for compute and bought some from a competitor
27. 87.▲
    by throwa356262 2mo ago

    This is laughable security. People claim security is now "solved" thanks to AI but from where I am standings it looks more like the fun 90ies making a return. Anyway, can someone explain the part about K3? What are they trying to
28. 88.▲
    by throwa356262 2mo ago

    Didn't we see this with Fable 5 on multiple benchmarks?
29. 89.▲
    by throwa356262 2mo ago

    Is the night window documented anywhere? Alibaba cloud website is almost as useless as their US counterpart (AWS). It's full of information everywhere but never what you need
30. 90.▲
    by throwa356262 2mo ago

    FYI: There have been reports of much smaller qwen derivatives being used for 0day research.