Posts on X before Jan 30, 2025, 1:28:26 AM UTC
I tried Trae (lmao) out, and I uninstalled it after 3 minutes. That's how bad it is
so it's true, this week is going to be model after model after model release-wise. yay!
yknow which models i never try out? the google ones. not until they're out of experimental. why? because through openrouter, they have an updatime percentage always in the low 30%. why offer it if it's unstable? let me pay for it and try the experimental ones out ffs
yeah honestly this Operator thing is amazing, I no longer have to sift through API documentation or dump it all into a huge context window hoping that it won't hallucinate an answer; now i just create a pastebin for it :)
PLEASE let me wake up to an hour-long (or however in-depth length I want it, make it set-able) X podcast based on everyone I follow
This would be AMAZING.
This confirms that what OpenAI has, it blows everything/everyone out of the water when it comes to showcasing capabilities, including Elon’s xAI.
Otherwise they’d have done this with xAI.
this is the first time a deepseek model feels consistently good for me; even deepseek v3 was "good", not GOOD. it had a lot of moments of stupidity, like the copy-cat it is. r1 is the first time it's consistently giving amazing results with little correction.
yeah so i didn't think i'd ever say this but i prefer deepseek's r1 vs. anything else, except for o1 pro. it's replaced all my needs that sonnet 3.6, gpt 4o, or o1 were satisfying before. all for a fraction of the cost. and it's sooooo malleable...
o1 pro + deepseek's r1 just 1-shots every single programming feat I throw at it. doesn't matter how many steps or how rambling my voice recording to it is regarding what i want. it just gets it.
another day, another piece of history. finally free.
Holy fuck. At last. Thank God.
one thing that really helps me deal with stupidity is remembering my own moments of stupidity
waiting for that sweet @openrouter nectar of providing API access to the new @deepseek_ai models
you think the job market is competitive and globalized now? give it a few more years ;)
we’ve finally hit january 20, 47 is stepping in, and talk of 200 executive orders is everywhere. no idea if that’s legit, but the change feels massive. i see it as a good thing overall. might rattle those who’ve been living in their own delusional echo chambers though
don't mind deepseek, they're just MOGGING every category
i must say i've been extremely impressed with how chatgpt is remembering past conversations to extremely fine minute details. i'm not sure how they're doing it, but whatever it is, it's working incredibly.
your wpm is keystroke based, my wpm is how fast i can babble coherently enough for whisper-large-v3 to understand me
if you're not coding with cursor like:
"do phase 1, then phase 3, then phase 4; repeat phase 3 until phase 4 returns successful"
yngmi
i'll be honest, sometimes i tweet something and then i look back at it in about 2-3 minutes and then i think to myself, damn, that was unnecessarily mean, and then i delete it
amazing how even the SotA models still recommend using hashtags for tweets, even with their 2023 knowledge cutoffs, they should know better no?
yeah the next update is super weird, i tried feeding the new openai ToS into itself as a PDF, and all chatgpt said was "it's good bro, just click accept :)"
weird but it helped me make lasagna from scratch last night so i trust it
anon, you know that you can now make what took a year, in a week?
Not limited to software btw, however ESPECIALLY in software
Literally nail biting and sleepless nights (on both political sides) at the potential presidency and future of our country / the world, immediately memory holed and talking about TikTok being banned soon LMAO
yeah i decided i'll never work for someone (other than my community) again after I dicked around back and forth for 2 months with a company I really connected with regarding their mission statement in the end after passing all their interviews and tests, they just emailed me (ba… Less
yeah i decided i'll never work for someone (other than my community) again after I dicked around back and forth for 2 months with a company I really connected with regarding their mission statement
in the end after passing all their interviews and tests, they just emailed me (basically) "nah, not a fit for now, maybe later one day?" LMAO
uh guys what the fuck is going on, this level of quality matched with it being literally free isn't supposed to be here until like, late 2025 at least. new year just started and we're already off with a banger here?
already have a goal for 2026: own one of these in Q1
lets see what the quality is like of this new chatgpt tasks stuff
unfortunately, text is becoming more and more noise and less and less high signal.
you really have to curate a good list with a track record before the prevalence of LLMs in order to have any chance of high quality.
everything else is slop, and the slop will only increase.
I feel like we're venturing into a period of time where text posts are just never going to hit the same. Unless you're reading something from someone with an established legacy, someone whose writings before LLMs were top-notch, you'll just never know nowadays whether what you’re… Less
I feel like we're venturing into a period of time where text posts are just never going to hit the same. Unless you're reading something from someone with an established legacy, someone whose writings before LLMs were top-notch, you'll just never know nowadays whether what you’re reading on Twitter, no matter how niche it is, was written by an LLM or by them.
There's this era of uncertainty and illegitimacy, and I think we're really going to feel it super soon. Especially with the next few major LLM upgrades, the speech patterns, the ability to stay on context, the recursive betterment aimed at truly making it seem like it’s not a bot reply, all of this is going to keep getting better and better.
this year i’m locked in on hitting 1m arr. top priority.
everything i learned last year i’m applying now, and i know it’s doable. if i can hit it, so can you.
the only time i’m typing is when i’m fixing my speech-to-text app. always tweaking it to fit my workflow, which changes daily.
crazy how much more productive you get when wpm isn’t the bottleneck. just speak your mind instead of burning headspace on typing.
After a lot of testing, this new Phi 4 model is cute and all, but anything with limited context windows like 16K is just pretty much unusable at this point, other than, I don't know just quick back and forth chatting that is short in duration
China has until the end of this month to release some bangers in AI in preparation for their upcoming silence during Chinese New Year. Looking at @Alibaba_Qwen @deepseek_ai mostly
Thinking back to Zuckerberg's recent statement and policy change, I think the next few years are going to be many-fold more powerful in terms of political and public change all across the world, more than we've already experienced so far, thanks to X. I think certain groups are … Less
Thinking back to Zuckerberg's recent statement and policy change, I think the next few years are going to be many-fold more powerful in terms of political and public change all across the world, more than we've already experienced so far, thanks to X.
I think certain groups are in for a rude awakening, and then other groups are in for the legitimization of their thoughts and will no longer feel like the odd one out simply because they never saw their own feelings replicated in the public forum.
Trust me, they will now.
Imagine in 4 years, all of North America is America 🇺🇸 lmao
if I see a — in your tweet, I stop reading and block you
lol, lmao even!
It really seems from what I'm reading that no matter what Zuckerberg does, his image is forever tainted and no amount of goodwill now will bring him back into positive opinion.
Which I disagree with for what it's worth, but that just seems how it is.
day 7/365 - completely reworked model caching and added favorites. instant loading, drag & drop organization, everything exactly where you want it
watch how it works ⬇️
day 6/365 - added custom provider support to systemsculpt. one click to set up local AI models
watch me go from zero to chat in 60 seconds ⬇️
who would have thought that on mediums with little to no moderation regarding freedom of speech, these would become the best and most accurate long-term forms of ideas?
i mean... does anyone have more data about the world than Alphabet? youtube, maps street view, GOOGLE (the search engine)...
it really seems like Google will win this all in the end unless someone beats them to it, which seems less and less likely each day
day 5/365 - making model selection smarter. organized by pricing tiers, sorted by capabilities, with quick visual indicators for what each model can do
plus: first alpha drops jan 8 for patrons
watch how it works ⬇️
day 4/365 - built audio transcription that's so meta i fed yesterday's video back into systemsculpt to analyze its own explanation. right-click any audio file, instant transcription + chat
watch how it works ⬇️
when API for @superwhisper ? @neilsuperduper
dev log day 3/365 - built custom ai personalities. turned a general ai into an ml research specialist that actually understands neural architectures
watch the before/after difference 🧵
just added instant context to systemsculpt - right-click any file and start chatting. pdfs, audio, images, even entire folders. handles everything automatically
day 2/365 ⬇️
hello @TheGregYang !
arrow key movement for images throughout a chat, as well as D for Download on the currently opened image, would be amazing adds!
