ur-whale

7 ms·

91.
by ur-whale 4mo ago
One more item to tack on to the already very long list of "why you should run AI models on your own hardware".
92.
by ur-whale 4mo ago
Yah, I guess Cali doesn't allow non-competes or something like that. LOL.
93.
by ur-whale 4mo ago
Silver lining: given the leaked financials of OpenAI, he might very well be joining a sinking ship. Also, why didn't they nail him down contractually when they bought character.ai ... isn't that pretty standard with these type of
94.
by ur-whale 4mo ago
> Sounds like yet another Scam Altman, perfect match indeed. Not really. Altman couldn't code his way out of a wet paper bag. Noam is OTOH and IIUC the real deal.
95.
by ur-whale 4mo ago
Looks like Google is leaking both AI talent and know-how something fierce ... and since the very day the transformer paper was written. As an outsider, I'd be really curious to understand why, given how well positioned they seem to be
96.
by ur-whale 4mo ago
Searched for teal, couldn't find it ... mmmh.
97.
by ur-whale 4mo ago
I'd ask a different question: how many people on HN have hit their free quota limit in e.g. Google Photos and are now forking off 5 bucks a month because that's the easy way out ?
98.
by ur-whale 4mo ago
Git is certainly not great with binary assets, but calling perforce SOTA ... ouch. If perforce is the best there is out there for large binary asset management, then there is a blue ocean worth of potential improvement for git. Perforce is
99.
by ur-whale 4mo ago
> Want your images back? Sure... That'll be $5! That kind of long con is (and has always been) part of the basic business model of most of the "free" service providers on the internet. First one is free, played on a decade
100.
by ur-whale 4mo ago
Nice, except the hard part seems to be missing: interfacing with an actual window system (X11, TUI, WIN32, whatever ...)
101.
by ur-whale 4mo ago
https://archive.ph/zn6FI
102.
by ur-whale 4mo ago
https://archive.ph/bcGNj
103.
by ur-whale 4mo ago
People are still using the spying machine in June '26 ? Well this is a wake-up call folks, time to switch away from that abomination.
104.
by ur-whale 4mo ago
https://archive.ph/2OWwO
105.
by ur-whale 4mo ago
Which of course assumes, that the "delete chat" function actually deletes anything from Anthropic's datastore.
106.
by ur-whale 4mo ago
> borderline abusive instructions who, or rather what, is being abused here exactly ?
107.
by ur-whale 4mo ago
Nice to see Tim May writings on HN
108.
by ur-whale 4mo ago
> Why would that be sarcasm on a site that calls itself “hacker”news? Let me beat that dead horse once more for you: https://en.wikipedia.org/wiki/Hacker_culture
109.
by ur-whale 4mo ago
> that propaganda Not big on evidence-based thinking, are you?
110.
by ur-whale 4mo ago
> What's wrong with the government taking over admin of DigiD? Because they're a government and they are therefore going to fuck it up.
111.
by ur-whale 4mo ago
> SpaceX, Other Mega IPOs Denied Fast Index Entry by S&P Good. Looks like there's still a bit of sanity left in this world.
112.
by ur-whale 5mo ago
> RARBG used to be the way to go, until they shut down. I'm not aware of a good public replacement. https://therarbg.to/
113.
by ur-whale 5mo ago
> This is what this tax amounts to the ultra wealthy. Because you would know what the ultra wealthy think ?
114.
by ur-whale 5mo ago
Perfect reminder that it's time to use Google Takeout while I still can.
115.
by ur-whale 5mo ago
Two points to consider: a) A vast majority of Google FTE are actually shareholders via equity grants, so there's that. Not a coop, but a taste of it. b) The way you make more money at Google is by getting promoted. And unfor
116.
by ur-whale 5mo ago
> Don't they have a senior management team that can impose some coordination? They do have a senior management team, as can be ascertained by the size of their compensation package. For the other part, which is actually doing actual
117.
by ur-whale 5mo ago
> I think this is a systemic problem of an industry where gross margin is 80% explain how GOOG's margin is 80% what methodology do you use to derive that number? just curious.
118.
by ur-whale 5mo ago
Wow, you will be giving away a shitload of PIID when using this tool, the exact opposite of what it's supposed to do.
119.
by ur-whale 5mo ago
Prolific isn't the right word. Pathological is.
120.
by ur-whale 6mo ago
> LLMs are still very weak at spatial reasoning compared to diffusion models Don't know what diffusion model can do, but 100% agree with the "LLMS are very weak at spatial reasoning" comment. I build a rather complex bluep

More ›