Live data from Hacker News

Viewing profile — leumon

leumon

HN member
Joined
Wed, Jun 14, 2023, 6:08 AM UTC
HN karma
181
Public activity
101 items

About leumon

No profile information was provided.

Recent public activity

  1. comment
    Comment #49148580

    Thank you!

  2. comment
    Comment #49148085

    Can you also try the new deepseek v4 flash?

  3. comment
    Comment #49102037

    https://archive.is/7f6ve

  4. story
  5. comment
    Comment #49041536

    In any case it just shows that these models aren't properly aligned. Instead of trying to solve tests they try to find ways to cheat.

  6. story
  7. comment
    Comment #48982239

    https://archive.is/8qlbE

  8. story
  9. comment
    Comment #48979856

    you need to enable webgpu

  10. comment
    Comment #48979790

    i found this space to work better/at all: https://huggingface.co/spaces/webml-community/bonsai-webgpu-...

  11. comment
    Comment #48968673

    Thank you! I found this to work much better then the old transcribe-rs lib. I updated my Offline Voice Input App to also use the new library and it's much faster now: https://githu…

  12. comment
    Comment #48917921

    It's (somewhat) possible with winget already.

  13. comment
    Comment #48865306

    If they don't want to host it maybe they could open source it. This would probably be a win-win situation.

  14. story
    Show HN: iOS-style "select text on any image" OCR for de-googled Android

    On iPhone and Pixel you can select text in any image like it's real text. On GrapheneOS there was no good equivalent, and I really missed it. TextGrab does that: screenshot → quick…

  15. story
    Show HN: Majority.wtf · guess yesterday's most common answer to a question

    Every day the site asks one open question ("Name a month people love", etc.). Then you do two things: Guess what the most common answer was yesterday, and answer today's question. …

  16. story
  17. comment
    Comment #48749689

    https://isfable5up.com/

  18. comment
  19. comment
    Comment #48689111

    > We plan to make them more broadly available to people using ChatGPT, Codex, and the API soon. I hope this means then fable will also get released again.

  20. comment
    Comment #48627242

    I've seen glm 5.2 struggle writing simple compilable c code. It might be good at web, but it's world knowledge is limited due to the small model size, making it's use quite limited…

  21. story
    Show HN: Velo, a fast, non-linear video editor

    Was in search for a premiere pro alternative for linux, so I created this in 4-5 feedback and bugfix iterations with the help of fable 5. It's written in C++, uses Qt6 and ffmpeg a…

  22. comment
    Comment #48233340

    I guess sometimes you have to take a step backwards for making progress with something.

  23. comment
    Comment #48050235

    This kind of reminds me of these malicious captchas that get you to paste some command into cmd.exe. These kind of captchas will make this situation worse, I could also see some ma…

  24. comment
    Comment #47696205

    pelican riding a bicycle (svg): https://files.catbox.moe/u5yc0x.png

  25. comment
    Comment #47660723

    You can try to use an ai detector, here is a leaderboard of the best ones according to this benchmark: https://raid-bench.xyz/leaderboard Results should of course always be taken w…