nuz

10 ms·

1. 91.▲
   by nuz 3y ago

   All those benchmarks are sooo bad that I'm not really convinced one way or another. Having tried out a few of them I roughly prefer gpt-4 I think still (though I haven't worked with any of them long enough to make a clear judgemen
2. 92.▲
   by nuz 3y ago

   All the thumbnail images are clearly dalle-3 too so there's more evidence https://scicoding.com/author/jake-2/
3. 93.▲
   by nuz 3y ago

   SDXL turbo is great in my experience! Way better than any alternative at that speed (e.g. sd1.4 or sd2). For img2img it's fantastic. And since the 800m model is using new insights since then and generally a cleaner dataset from the loo
4. 94.▲
   by nuz 3y ago

   Imagine how many hundreds of millions of users waste their time using instagram and facebook on a daily basis. Safe to say they don't mind wasting their customers time
5. 95.▲
   by nuz 3y ago

   HN also seems unusually slow. Wonder what's going on
6. 96.▲
   by nuz 3y ago

   The 800m model is super exciting
7. 97.▲
   by nuz 3y ago

   Both SD1.4 and SDXL was in limited preview for a few months before a public release. This has been their normal course of business for about 2 years now (since founding). They just do this to improve the weights via a beta test with less ju
8. 98.▲
   by nuz 3y ago

   More opportunity for mistral and other EU competitors then I suppose
9. 99.▲
   by nuz 3y ago

   Crazy to be so ahead of the curve but sacrifice all first mover advantage in an entire continent like this.
10. 100.▲
    by nuz 3y ago

    GPT-4 was created like 3 years ago internally
11. 101.▲
    by nuz 3y ago

    Anyone understand what the point no 5 means? What regulation letter is it referring to?
12. 102.▲
    by nuz 3y ago

    > ...part-time partners (we do that with Peter Thiei for exampie, though at this point he's very involved) From the quoted text above. I.e. more than part time partner (heavily involved)
13. 103.▲
    by nuz 3y ago

    Yet interestingly kept donating to openai until september 2020 according to the document.
14. 104.▲
    by nuz 3y ago

    New? Looks the same as it has been for years. Also sam is clearly speaking in general terms.
15. 105.▲
    by nuz 3y ago

    > I'd hate to bet against elon winning - @sama https://twitter.com/sama/status/618265660477452288
16. 106.▲
    by nuz 3y ago

    Personally all the fluid simulation shaders I've written usually makes my fan go off, and I'm counting a few of those here so that's impressive in my eyes.
17. 107.▲
    by nuz 3y ago

    Blows my mind that my computer isn't even turning on its fans considering how many shaders are running in this thing.
18. 108.▲
    by nuz 3y ago

    Absolutely beautiful article and presentation
19. 109.▲
    by nuz 3y ago

    He's probably right. But the upside is this: Think about all the hundreds of little coding ideas (games etc) that you have in your brain or written down somewhere that would have taken years to develop and millions in opportunity cost
20. 110.▲
    by nuz 3y ago

    Fine by me. They have to get money somehow so this is expected, and in return we get top notch models to use for free. I don't mind it.
21. 111.▲
    by nuz 3y ago

    Hallucinations are unintended. These are intended and built into the model very consciously
22. 112.▲
    by nuz 3y ago

    I'm not even seeing any people on my version. Just devices. Wonder why
23. 113.▲
    by nuz 3y ago

    GPT-4 was released at the same time as Bard was announced I believe (same day, same hour basically).
24. 114.▲
    by nuz 3y ago

    I saw similar artifacts in dalle-1 a lot (as if the image was pasted onto geometry). Definitely wouldn't surprise me if they use synthetic rasterized data to in the training, which could totally create artifacts like this.
25. 115.▲
    by nuz 3y ago

    This is the second time OpenAI has released something right at the same time as google did (Gemini 1.5 Pro with 10M token context length just now). Can't just be a coincidence
26. 116.▲
    by nuz 3y ago

    AGI at the quality of sora or dalle but for intelligence is gonna be quite the thing to witness
27. 117.▲
    by nuz 3y ago

    Short on money after being an exec at tesla during a huge rise in its stocks? More likely he has too much money and maybe doesn't really want or need to work and is doing passion projects instead
28. 118.▲
    by nuz 3y ago

    I imagine he's talking about his youtube series (look forward to it!)
29. 119.▲
    by nuz 3y ago

    Gentrification is a poor analogy since the web is not zero sum unlike physical space. Attention is limited though and that has shifted away from websites like this so maybe it's gentrification of attention
30. 120.▲
    by nuz 3y ago

    So many scroll hijacking ones...