Microsoft Phi-4 (14B) : This Opensource LLM is a MINI BEAST! The Best 14B Model YET! (Beats Qwen!)

Поделиться
HTML-код
  • Опубликовано: 19 дек 2024

Комментарии •

  • @wmzayed
    @wmzayed 2 дня назад +50

    I believe it is time to change your 13 test questions. I feel the Microsoft PHI team is following you and training the model around your questions. :). You can create a different set of questions similar in concept.

    • @kafkaesqued
      @kafkaesqued 2 дня назад

      😂😂

    • @luismoriguerra669
      @luismoriguerra669 2 дня назад

      hahha classic benchmark issue

    • @HemangJoshi
      @HemangJoshi 2 дня назад +1

      Actually the best benchmark is aider leadboard. Whichever LLM is on top it is the best period.

  • @You12783
    @You12783 2 дня назад +37

    You should make a longer video by creating a full stack application using the models who've scored really good in your benchmarking questions. That way we'll know which one's the best.

    • @aculz
      @aculz 2 дня назад +2

      then join the membership, simple

    • @midzuushi
      @midzuushi 2 дня назад

      ​@@aculz thx for the Info🎉

  • @EditUMedia
    @EditUMedia 2 дня назад +7

    My only concern is the model only being good on benchmarking questions, because of the history of Phi models being trained specifically to score high in benchmarks rather than real world performance.
    But this model seems promising, I'm excited to try it out.

  • @trokk24
    @trokk24 2 дня назад +1

    Next Level and local. I expect that tool use becomes a feature. It would greatly enhance the potential. I've ran it at q8, q6, and q4 and basically got the same performance. Trying it now with the settings as you recommended. Thanks for sharing CodeKing.

  • @paulyflynn
    @paulyflynn 2 дня назад +7

    Can you do simple, non-impactful changes to the questions? for example, "2 plums" instead of "2 apples"

  • @themarksmith
    @themarksmith 2 дня назад

    excellent video dude!!!!

  • @rrioclkls7721
    @rrioclkls7721 2 дня назад

    Does open web ui normally display generated pages (like with the confetti button @ ~9:11)?

  • @AB-cd5gd
    @AB-cd5gd 2 дня назад

    Best test is asking for a modern sleek landing page, you quickly see how good or bad the model is

  • @_lun4r_
    @_lun4r_ 2 дня назад +2

    can't wait for Phi-4 small (~7B) and Phi-4 mini (~3B) and make it crush all benchmarks in these ranges
    the Phi-4 you're showcasing here is a Phi-4 medium

  • @VietVuHunzter
    @VietVuHunzter 2 дня назад +10

    Lmao I won't trust Phi models until real world benchmark like arena/live bench.

    • @tukanhamen
      @tukanhamen 2 дня назад

      Yep been disappointed too many times

  • @PseudoProphet
    @PseudoProphet 2 дня назад +1

    You can run it on even a M4 Mac mini .

  • @jimlynch9390
    @jimlynch9390 2 дня назад

    That's a quite good model. Thanks.

  • @Tyrexxllc
    @Tyrexxllc 2 дня назад

    I think it's time to update your test questions!!

  • @kydjester
    @kydjester 2 дня назад

    next time paste all the questions at once and lets see the fun.

  • @SipChai
    @SipChai 2 дня назад

    Can you compare other small models?

  • @midzuushi
    @midzuushi 2 дня назад

    Open Ai ... its gonna be Open

  • @Adam-fl9uc
    @Adam-fl9uc 2 дня назад

    Woooooooow! It is incredible

  • @njt4u
    @njt4u 2 дня назад

    This is quite insane 😮

  • @1-chaz-1
    @1-chaz-1 2 дня назад

    Wow

  • @다루루
    @다루루 2 дня назад

    🐿️🐿️🐿️🐿️🐿️🐿️