mixtureoftakes

6 ms·

31.
by mixtureoftakes 6mo ago
what is the main difference between "normal" quants and the UD ones?
32.
by mixtureoftakes 7mo ago
github unavailable; what you think would be the primary usecase for agent to agent chat?
33.
by mixtureoftakes 7mo ago
Ironically, author doesnt realize that not everyone wants to learn everything There is nothing wrong whatsoever with just getting things done.
34.
by mixtureoftakes 7mo ago
happened to my friend too! an overkill but working solution for this is "sudo pmset -a disablesleep 1"
35.
by mixtureoftakes 7mo ago
i think its all about caring and knowing what you want to make and willing to iterate on the result until it is actually good. If you want the ai to do your job for you its probably not going to work, but if youre really good at using its a
36.
by mixtureoftakes 7mo ago
Yeah and also they still want to get at least some sales on the mac studios and mac pros with ultra chips, 256gb m5 max wouldve straight up killed both of those products. We can have nice things but nobody is going to hurt themselves to giv
37.
by mixtureoftakes 7mo ago
laptops seem kinda solved now? for every cpu vendor and every os theres now a great option that just, works, so everyone can just use what they like and not be at a disadvantage due to not knowing the latest developements in the space, or d
38.
by mixtureoftakes 7mo ago
You'll get slightly more performance and ever so slightly less battery life. I'd do it
39.
by mixtureoftakes 7mo ago
I mean, they had to make them good because of the new cpu architecture, but since the emulation worked so well and overall adoption was really fast it now is a problem for them as a company. A really good problem to have though
40.
by mixtureoftakes 7mo ago
Curious about that as well. They seem to market it as a technological advancement, which it is, but rather than being excited im actually worried about hidden latencies that could come with that approach. Have you found any interesting inf
41.
by mixtureoftakes 8mo ago
media.tenor.com/oW5zO_6gu5gAAAAi/theomegaoof-emoji.gif
42.
by mixtureoftakes 8mo ago
what is with the ai written articles lately? and ai designed websites? who looks at it and goes "yes, ill upvote this. more people should see it."?
43.
by mixtureoftakes 8mo ago
not only is it down but its likely to be a decently sized hack too, many big accounts including elon having their bio edited to promote various crypto
44.
by mixtureoftakes 8mo ago
the comment " be aware that talking to LLM actually moves carbon from earth into atmosphere" having 39 likes is ABSURD to me. out of all the fascinating and awful things to care about with the advent of ai people pick co2 emission
45.
by mixtureoftakes 8mo ago
i am impressed. no ai can ever write announements this bad
46.
by mixtureoftakes 8mo ago
extremely bad takes CHECK broken cert CHECK somehow top of front page CHECK
47.
by mixtureoftakes 9mo ago
also is there a way to make parakeet type more naturally? less capitallization, less punctuation? can this be a setting? this can already be done via local llm processing the text but surely there is an easier way to do this, right
48.
by mixtureoftakes 9mo ago
Which model would be the best to use for mandarin? Are there any models on par with Parakeet that are just as fast but also understand Chinese?
49.
by mixtureoftakes 9mo ago
the model is permanently loaded into ram for access speed. discharging it would unload it from ram and lead to longer start times
50.
by mixtureoftakes 9mo ago
oh this is because they want more data to build better ai (that will give them more money and power and probably some other things too)
51.
by mixtureoftakes 1y ago
so you have ptsd from a browser?
52.
by mixtureoftakes 1y ago
WHY IS EVERYONE IN COMMENTS SO OLD AND BITTER WHAT ON EARTH ARE WE DOING
53.
by mixtureoftakes 2y ago
15 years in tech is like a month in crypto
54.
by mixtureoftakes 2y ago
this is crazy when trying to run on a mac it only plays in a very small window, how could this be configured?
55.
by mixtureoftakes 2y ago
small world we live in...
56.
by mixtureoftakes 2y ago
fuzzing lore: https://threadreaderapp.com/thread/1799457232607985698 great read if you wanna waste 11 minutes
57.
by mixtureoftakes 2y ago
Is this sota? How does rag quality compare to that in, say, lmstudio? Also is there a "rag quality benchmark" yet
58.
by mixtureoftakes 2y ago
Most of us will read this and continue living our life exactly the same way as before …wake up
59.
by mixtureoftakes 2y ago
TLDR - 2x cheaper, slightly smarter, and they only compare those new models to their own old ones. Does google have moat?
60.
by mixtureoftakes 2y ago
Honestly i would have typed commands in shell if "captcha" asked me for it. Just to see the scale of outcome's awfulness. I'm almost bored enough to just start installing weird malware for research and funsies

More ›