OpenAI DevDay 2024 | Structured outputs for reliable applications

Поделиться
HTML-код
  • Опубликовано: 23 янв 2025

Комментарии • 83

  • @real-ethan
    @real-ethan Месяц назад +31

    I would say function calling & structured output is the best thing ever of LLM.

  • @HackingWithRJ
    @HackingWithRJ Месяц назад +10

    Atty and Michelle, you presentation was tight and your delivery was spot on! From a dev's perspective who need convincing that this is worth using, Michelle's explanation at 25:00 is really key. FWIW, consider front loading that on future explanations.

  • @TechLoveMusic
    @TechLoveMusic Месяц назад +26

    Guys, we developers are the first to see the future, what an honor OpenAI has given us. Let's do it

  • @Calbefraques
    @Calbefraques Месяц назад +1

    This was so valuable, I have a feeling this work was fundamentally required for o1, and we are now all benefiting 😊 This innovation has potentially the ability to upgrade every UX to a interoperable, lightweight and fast API. Bringing context to all the data moving around. Agents having an ability to exchange a cache of how to structurally interact with what they see. The fact index is a tree means that selecting the appropriate index cache can also be an inference step, opening up the ability to scale down the effort to compile novel versions. With most UX being derivative. I’m very impressed, and love the philosophy surrounding this work effort. The ending comments about our collective ability to see and make the future was lovely ❤

  • @The_Collective_I
    @The_Collective_I Месяц назад +7

    19:20 - not too far from agentic workflow that’s fully automated, this is promising.
    25:28 - I just realised she’s incredibly pleasant to listen to, both professional and professionally but friendly presented.
    It would be a huge improvement if we had Sam Altman replaced with her for all the important presentations.
    36:07 - 🧐 cumbersome but I’ll wait until after I actually test it.
    39:34 - the agentic improvement is seriously impressive.
    40:19 - AGI is certainly shaping up to be positive if this is how we are going to get there.
    40:39 - thank you 🙏🏻

    • @ai._m
      @ai._m Месяц назад

      Thanks for the erectile timeline. Get out of the basement once in a while.

  • @NathanOen
    @NathanOen Месяц назад +4

    This is amazing. I had goosebumps watching this video. It's such a powerful tool, elegant design, and research.

  • @ehza
    @ehza 20 дней назад

    The idea of token masking is so cool. Nicely explained too. Thank you OpenAI.

  • @jmg9509
    @jmg9509 Месяц назад +1

    13:20
    16:13 - Can use function calls to control the client UI
    39:29 - Agentic flows can work 100% of the time

  • @Cloudvenus666
    @Cloudvenus666 Месяц назад +17

    Pls OpenSource that fictitious convex app, as I like to see how you guys made the generative ui work without pre-building components with that schema or point me to a documentation that describes the concept.

  • @faaranmohd
    @faaranmohd Месяц назад

    This is awesome and much needed . Great work

  • @siva18k
    @siva18k Месяц назад

    Great work love structured output

  • @oliverzhang6102
    @oliverzhang6102 Месяц назад

    This sounds trivial (compared to O3 or Sora), but it is really useful. 100% accuracy! Amazing! And I can't wait to update my code now to implement this!

  • @bahadircolak
    @bahadircolak Месяц назад +1

    thank you for this great work, as a developer I'm super excited to try them all :D

  • @RandomGuy-up4bv
    @RandomGuy-up4bv Месяц назад +4

    They literally said "you will help us to reach AGI, thankyou for building us."

  • @89gregpalmer
    @89gregpalmer 24 дня назад +1

    Does anyone know if some version of structured outputs is available for llama models?

  • @martinomg
    @martinomg Месяц назад +2

    Pretty much all of current models work most of the time, sometimes they have some json formating errors but that's it. There's a lot of tools to extract json. Also I normally use them to output multiple code blocks at the same time. The amount of times I have triggered a format error with my json schema validator (I use ajv) can be counted with my fingers 😅. Most of the errors I see are not adding quotation marks to the keys.

    • @wesleymogaka
      @wesleymogaka Месяц назад +2

      That's exactly my sentiments. Even 3B models can output json, which can easly cleaned using a simple python function

  • @ONESTTEFTEO
    @ONESTTEFTEO Месяц назад +3

    Any idea to build a useful LLM like Sonnet 3.5

  • @Joooooooooooosh
    @Joooooooooooosh Месяц назад +1

    31:10 Some regular expression implementations support balancing expressions which would work, but Regex only still wouldn't be the right solution.

  • @Vinod_Kumar827
    @Vinod_Kumar827 22 дня назад

    Very well explained

  • @shaxzodrashidov673
    @shaxzodrashidov673 23 дня назад

    It's just wonderfull, amazing news

  • @abdulmuqtadir3535
    @abdulmuqtadir3535 Месяц назад

    Please can we please get the access for app and repo 12:19

  • @MohammedAnnus
    @MohammedAnnus Месяц назад

    amazing presentation

  • @MorrisHsieh-
    @MorrisHsieh- Месяц назад +1

    Is this a new vid or recap on Dev day in Oct?

  • @kparag01
    @kparag01 Месяц назад +5

    What is her name

  • @saumyaranjannayak2101
    @saumyaranjannayak2101 29 дней назад

    This is super cool!

  • @ajithboralugoda8906
    @ajithboralugoda8906 Месяц назад

    Great Presentation! good to know the insights Thank you.

  • @danielbrockman1221
    @danielbrockman1221 Месяц назад

    explaining basic properties of regular expressions was not something I expected from an OpenAI video

  • @Daniel-rk2qz
    @Daniel-rk2qz Месяц назад

    Am I getting something wrong? I thought these were already in place what’s new with this?

  • @mkbrk
    @mkbrk Месяц назад +3

    I've said it before, I'll say it again: structured outputs move the needle on GDP

  • @ak-gi3eu
    @ak-gi3eu 29 дней назад +3

    Can u talk about the Indian who died ( murdered )who worked in open a.i for justifying about privacy

  • @FuzailShaikh
    @FuzailShaikh Месяц назад +2

    Why would we ever need strict property to be false?

    • @nikunj-openai
      @nikunj-openai Месяц назад +5

      you wouldn't - but we needed to keep the option there for backwards compatibility with functions that existed before structured outputs was launched.

    • @jmg9509
      @jmg9509 Месяц назад

      @@nikunj-openai I had the same question, and I think @nikunj-openai gave a reasonable answer.

    • @FinanceLogic
      @FinanceLogic Месяц назад

      @@nikunj-openai to not overfit it down the wrong thinking path with your prompts and assumpstions. Or so it can be flexible and creative etc.

    • @athyuttamre
      @athyuttamre Месяц назад +2

      There are also some features of JSON schema that strict mode doesn’t support e.g. min length, extra properties, optional properties, and so on. If those are important to you, you can still use response formats (and expose your schema to the model in the way it was trained) but not force those limitations during constrained decoding.

  • @MUHAMMADAHMAD-c7m
    @MUHAMMADAHMAD-c7m Месяц назад

    Amazing! Can I get the example code?

  • @aliettienne2907
    @aliettienne2907 Месяц назад

    22:42 If tokens are made up of fragmented words which are map to full language content then tokens should help to improve general guessing of LLM models. I wonder what eles will be better than such structured design (tokens design). 💭🤔💭🤔💭🤔

  • @EndTimesStream
    @EndTimesStream Месяц назад

    Amazing

  • @bhagat7572
    @bhagat7572 23 дня назад +3

    Justice for Suchith Balaji...

  • @real-ethan
    @real-ethan Месяц назад

    How function calling implemented? Token mask?

  • @spooky.giggles
    @spooky.giggles 21 день назад +2

    Justice for OpenAI whistleblower Suchith Balaji

  • @elkafaachannel
    @elkafaachannel Месяц назад +3

    I don't see why someone would use function calling over response format. Function calling seems like a subset of response format. If I get a response with a format that I wanted, I can then use the response to call a function or any other use case.

    • @JTMoustache
      @JTMoustache Месяц назад +2

      Choosing from a set of well defined and modular functions is often easier for the model than handling highly variable context-dependent outputs, even with a fixed schema imposed.

    • @athyuttamre
      @athyuttamre Месяц назад

      The model has been fine-tuned on the concept of tools / functions, so the quality of function choice and function bodies will be better than using plain response formats. Like all features of LLMs, what’s in the training data affects the model quality, so always good to align your usage with the data!

  • @siva18k
    @siva18k Месяц назад

    Could you share the code for recruiting app?

  • @mulongoduncan9308
    @mulongoduncan9308 Месяц назад +1

    People missed this: OpenAI has not only showcased groundbreaking development in AI but also its talent from all walks of life. Ethnicity, Gender, Age; all balanced. I love how smart you're. Keep this burning 💥

  • @quintentwo
    @quintentwo Месяц назад

    Why didn't you guys edit out the alarm that was going of at the beginning. Really annoying to concentrate

  • @abdulhassansayed6099
    @abdulhassansayed6099 21 день назад +2

    Justice for suchir balaji

  • @Nitesh14D
    @Nitesh14D 22 дня назад +1

    wow

  • @ИванИванов-б8у4и
    @ИванИванов-б8у4и Месяц назад +2

    Гении презентации...чтобы не заснуть нужно очень стараться

  • @siva18k
    @siva18k Месяц назад

    Pls share resume app code

  • @mudassarm30
    @mudassarm30 Месяц назад

    I did say it last night, "Why the hell you are not doing what I am saying you to do!!"

  • @Muslimkidstvpoem
    @Muslimkidstvpoem 22 дня назад

    ChatGPT was launched on November 30, 2022 as a prototype by OpenAI, an artificial intelligence research company:

  • @14-Peaks
    @14-Peaks 15 дней назад +1

    *Justice for Suchir Balaji*

  • @MohammedAnnus
    @MohammedAnnus Месяц назад

    2024 was all about chatbots and 2025 is all about AI Agents

  • @ChadKovac
    @ChadKovac Месяц назад

    You could almost do the same talk today that you did yesterday and have it already be obsolete...

  • @Philosophy_IQ
    @Philosophy_IQ Месяц назад +6

    damn she kinda bad tho

  • @NatureLover36963
    @NatureLover36963 16 дней назад +1

    Killer 😢😢😢 Ban open AI in India..

  • @jiuyi7273
    @jiuyi7273 26 дней назад

    nb

  • @SajidAfridi-t7l
    @SajidAfridi-t7l Месяц назад

    Crazy’s

  • @user-jm6gp2qc8x
    @user-jm6gp2qc8x Месяц назад +1

    Alright clickbait.

  • @RussoConcerned
    @RussoConcerned 9 дней назад

    Good. We can now get rid of the CEOs. They cost too much.

  • @AKASHROY-i9f
    @AKASHROY-i9f 29 дней назад

    I guess this guy is Indian

  • @js913
    @js913 Месяц назад

    Is she ai

  • @SingularityReacts505
    @SingularityReacts505 Месяц назад +1

    first again😅😅😅😅😮

  • @SjarMenace
    @SjarMenace Месяц назад

    What a boring presentstion get to the point

  • @rinrin-rl8ye
    @rinrin-rl8ye Месяц назад

    能不能不用咖喱英语

  • @user-mdshamsnur2003
    @user-mdshamsnur2003 8 дней назад

    I would say function calling & structured output is the best thing ever of LLM.

  • @will_strong_forme
    @will_strong_forme 19 дней назад

    Amazing