Live data from Hacker News

Viewing profile — villgax

villgax

HN member
Joined
Wed, Nov 14, 2018, 12:26 PM UTC
HN karma
1,002
Public activity
1,152 items

About villgax

SAY GOONER

Recent public activity

  1. comment
    Comment #49144874

    No mention of Infiniband or SPX lol

  2. comment
    Comment #49142108

    I went the gpus.io website & it’s $2.95/hr right now, this is like comparing MSRP to actual market price. B300s are in demand & hence cost more, but these lazy editors at wafer.ai …

  3. comment
    Comment #49141794

    Lol, such a lazily written article by wafer.ai GPUs. 8× MI355X (TP8) B300 (TP8+DCP8) Decode tok/s per stream 118 tok/s 172 tok/s Peak aggregate. 952 tok/s 1,568 tok/s Peak aggregat…

  4. comment
    Comment #49141097

    Checks notes, the same exact line of commentary that exists for Spacex themselves

  5. comment
    Comment #49141095

    Actually read the post before ignoramusing here

  6. comment
    Comment #49134544

    No wonder search in codex is so a$$

  7. comment
    Comment #49044399

    Even AI generated art needs taste, looks so bland & off-point.

  8. comment
    Comment #49044223

    I would not even go to the extent of pay to call, just pay to be let through, regardless of whether I pick up or call. Don’t even want to get into chargebacks coz I didnt pickup lo…

  9. comment
    Comment #49039513

    you can imagine if they bother cracking landing back to re-use, it's KO for spacex

  10. comment
    Comment #48919822

    This should be the case even for phones as well

  11. comment
    Comment #48906296

    @claude add this to the graveyard of wannabees

  12. comment
    Comment #48887302

    World’s first? Such lazy, much farming https://github.com/fla-org/native-sparse-attention?utm_sourc...

  13. comment
    Comment #48868941

    AirDrop you mean lol? Anyone can now have a local LLM make a QR code based data transfer script

  14. comment
    Comment #48800260

    All these names mean squat

  15. comment
    Comment #48782719

    They fail to mention non speculative numbers & whether baseline was nvfp4 as well. So much for erosion against an older gen

  16. comment
    Comment #48782711

    Lot more SMs & Tensor Cores for NVFP4 going by the looks of it.

  17. comment
    Comment #48769193

    Doesn’t even have to be merged, it’s for the privilege of even being considered to be merged into a popular library, one can always fork but to put stuff in a popular repo should b…

  18. comment
    Comment #48744063

    Just put it behind a paywall for PR prioritization or consideration, more payment to jump the queue. There, I solved FOSS sponsorship.

  19. comment
    Comment #48705172

    PRs just need a paywall with a priority tier tied again to money. If you want AI code to show up in a popular repo, pay for a chance then or just fork it for free.

  20. comment
    Comment #48663729

    Will it skip Ads lol

  21. comment
    Comment #48643415

    I mean he did successfully lead companies with barely any profit, amass huge customer profiles, selling to both a bank & a social media Ad tech giant. Not surprised at all.

  22. comment
    Comment #48554248

    Thank god ISRO isn’t run by the same group of people

  23. comment
  24. comment
    Comment #48517550

    Just switch to Bifrost already

  25. comment
    Comment #48487217

    This is an exhibit of implosion