jdthedisciple

4 ms·

31.
by jdthedisciple 3mo ago
I too have been wondering about the missing timestamps, and sometimes feel like it would be really useful. My best guess is fear of context pollution, but no idea...
32.
by jdthedisciple 3mo ago
> Not showing us Have you watched the 46s video full screen on a monitor and not marvelled at the incredible 4K detail of the FPV motorcycle racing clip?
33.
by jdthedisciple 3mo ago
It's incredible how negative and dismissive the comments in here are while here I am thinking the model actually looks impressively capable. But then again I heard the downers have always been the first to leave their dung comments her
34.
by jdthedisciple 3mo ago
None of those are SWE. And still Fable beats it hands down 8-0 in one of them, and is at worst even in some others. Also it doesn't make logical sense: A router can save costs, yes, but not magically be "smarter" somehow. T
35.
by jdthedisciple 3mo ago
Grandiose claims that are hardly backed up at all. I find that very off-turning!
36.
by jdthedisciple 3mo ago
Bottom line it looks about on equal footing with GLM 5.2 in terms of both overall intelligence and cost per task, while being significantly faster (in fact it is the fastest model on artificial analysis as of rn [0]) [0] https://
37.
by jdthedisciple 3mo ago
Interesting, well it'd be interesting to check out some individual examples where Gemini beat the others. Also, would be great if you could add GPT 5.6 Sol XHigh and Fable 5 High as well, just to see if at least those beat Gemini which
38.
by jdthedisciple 3mo ago
How does your comparison work? It places Gemini 3.6 Flash Medium above GPT 5.6 Sol High and Fable 5 Medium, which makes me skeptical because that... would be making headlines that I'm not seeing right now.
39.
by jdthedisciple 3mo ago
No, your thinking is 100% correct. It's called clickbait.
40.
by jdthedisciple 3mo ago
I recently had GPT 5.6 work on something for 54 minutes while I went to sleep
41.
by jdthedisciple 3mo ago
> In a few years there will be Mythos level open weight At the rate things are moving I'd expect that to happen much sooner. In fact: Somebody, right now as we speak, is most likely already working on training the next best open sou
42.
by jdthedisciple 3mo ago
> Kimi will train their models on your interactions I find these kinds of concerns increasingly silly: most of the input to these models will be ... previous output from the very same models, alongside the occasional half-assed human com
43.
by jdthedisciple 3mo ago
That explains it! I was wondering about this just today as I saw it hd reset. Alas I don't need to wait another 7 days..
44.
by jdthedisciple 3mo ago
Compelling dashboard, bookmarked to check in every now and then
45.
by jdthedisciple 3mo ago
This was my first thought too, and your comment the first one I saw when I clicked this. Just thought I'd share that..
46.
by jdthedisciple 3mo ago
Insane...
47.
by jdthedisciple 3mo ago
Impressive indeed!
48.
by jdthedisciple 3mo ago
> While its overall performance still trails the most powerful proprietary models, Claude Fable 5 and GPT 5.6 Sol I'm inclined to believe that, however according to their own benchmarks Kimi K3 actually even beats the other two in m
49.
by jdthedisciple 3mo ago
Easy mitigation: Disable memories, use fake name
50.
by jdthedisciple 3mo ago
> load-bearing Claude! [0] [0] https://news.ycombinator.com/item?id=48905248
51.
by jdthedisciple 3mo ago
The theory felt a bit oversold in the introduction, imho.
52.
by jdthedisciple 3mo ago
Anybody compared this against Whisper Turbo?
53.
by jdthedisciple 3mo ago
Quick feedback: I searched "n8n" and it did not surface SimStudio [0], which is pretty much an open source clone of it. [0] https://github.com/simstudioai/sim
54.
by jdthedisciple 3mo ago
this is the way
55.
Ask HN: Are current on-device LLMs useless?
1 points by jdthedisciple 3mo ago|1 comments
56.
by jdthedisciple 3mo ago
Open a Patreon page with different tiers and a private Discord server for your patreons. From what I gather that's been a fairly successful standard model for independent content creators.
57.
by jdthedisciple 3mo ago
I'm fine with the timer if at least the game isn't stopped after a fail. Just reduce the final score, as someone else here also suggested, but I'd like to at least always get to try all 18 words.
58.
by jdthedisciple 3mo ago
Devil's advocate: The widespread lack of social skills was not caused by LLMs in the first place. In fact, AI ChatBots, if used right, can actually help remedy lacking socialising skills.
59.
by jdthedisciple 3mo ago
You could also probably one-shot your own version of this using any SOTA LLM, it's not terribly complicated (but a nice game idea!).
60.
by jdthedisciple 3mo ago
So it also has good humor, kinda sounds like a cool friend no?

More ›