Topfi

55 ms·

451.
by Topfi 11mo ago
> [...] but it's not something they can do without being blatantly obvious. Thing is, it is being done by certain labs and it is blatantly obvious [0]. It has for months now been very easy to tell which labs either "train to th
452.
by Topfi 1y ago
Thanks for sharing. Honest question, where did you find this? Isn't linked in their status page or blog and I find it weird that OAI would share it like that when they usually link to PDFs or just make regular blogposts. Seems to be fr
453.
Closing Windows 11's Task Manager accidentally opens up more copies(arstechnica.com)
18 points by Topfi 1y ago|1 comments
454.
Reregister your hardware 2FA key for X before November 10(techcrunch.com)
1 points by Topfi 1y ago|0 comments
455.
MiniMax-M2: Intelligence, Performance and Price Analysis(artificialanalysis.ai)
3 points by Topfi 1y ago|1 comments
456.
by Topfi 1y ago
As a former AVP owner, this seems much more suitable for what I personally tend to use these devices for, with the headstrap appearing far closer to my personal comfort preference (halo straps) and a lower weight due to reduced power and no
457.
by Topfi 1y ago
> AIME is saturated (with tool use) [...] But isn't tool use kinda the crux here? Correct me if I'm mistaken, but wasn't the argument back then on whether LLMs could solve maths problems without e.g. writing python to solv
458.
by Topfi 1y ago
Guess even the CTO of OpenAI relies on Anthropic models in a pinch...
459.
by Topfi 1y ago
Reminds me of an ARG [0] I made in the early days of LLM hype. Honestly had three mails asking whether they could invest. Likely scams if we are honest, just some automated crawlers, but found it funny nonetheless. [0] https://et
460.
by Topfi 1y ago
If holding the CTO of OpenAI accountable for his wildly inaccurate statement constitutes "dunking on OpenAI", then I'd say dunk away. He, more than anyone else, should be able to for one parse the original statements correctl
461.
by Topfi 1y ago
> What good is your open problem set if really its a trivial "google search" away from being solved. Why are they not catching any blame here? They are a community run database, not the sole arbiter and source of this informati
462.
by Topfi 1y ago
Isn't the dot com bubble a far better proxy? Notably, todays spending is both higher and more concentrated in a few companies that a large part of the population has exposure to (most dot com companies weren't publicly traded and
463.
OpenAI researcher announced GPT-5 math breakthrough that never happened(the-decoder.com)
430 points by Topfi 1y ago|234 comments
464.
by Topfi 1y ago
Accutrons and tuning fork watches are amazing. They have an incredibly unique sound/hum due to the tuning fork oscillating at 360 hz and the most smooth glide you'll ever see in a watch. Recommend a ESA 9162 or ESA 9164 over a pur
465.
by Topfi 1y ago
Thanks, sorry, only saw the offer now. Have just checked and cannot reproduce the usage any more, might have been mistaken on that front.
466.
by Topfi 1y ago
It goes back to Sparks of AGI [0] unless I am mistaken. Can recommend the talk, one that has stayed in the back of my mind since I first saw it two years ago. Personally, still have major reservations about throwing claims of intelligence o
467.
by Topfi 1y ago
Gorilla is a great resource and it isn't unreasonable to suspect Z.AI has it in their data sets. I'd suspect most other frontier labs as well (pure speculation, but why not use it as a resource). Problem is, while Gorilla was an a
468.
by Topfi 1y ago
Has been ongoing for roughly a month now, with a variety of checkpoints along the usual speculation. As it stands, I'd just wait for the official announcement, prior to making any judgement. What their release plans are, whether a chec
469.
by Topfi 1y ago
I wish someone would answer your question. As it stands, without any example, I for one have no idea what is being talked about, where these popups in FF are, how they affect accessibility. Do I dislike modals that cover the entire window,
470.
by Topfi 1y ago
Could you link some? Would love to see what dialogs are meant, whether they can actually cause accessibility issues or are just "unwanted" and how the maintainers comment/decide on that front. Been a hot minute since I last i
471.
by Topfi 1y ago
Could have phrased that a bit better, but I did mean that while there are use cases in which the delta between Haiku, Sonnet, Opus or another providers model are clear, this is not the case for every task. In my experience, yes, Opus 4 and
472.
by Topfi 1y ago
Fair point of course and it is still far to early to make a definitive statement, but in my still limited experience throughout the night, I have seen Haiku 4.5 be far better in using what I'd consider a justifiable amount of input tok
473.
by Topfi 1y ago
As added context to ensure no benchmark gaming, here a quite impressive Shitaki Mushroom riding a rowboat: https://imgur.com/Mv4Pi6p Prompt: https://t3.chat/share/ptaadpg5n8 Claude 4.5 Haiku (Reasoning
474.
by Topfi 1y ago
Was just about to post that Haiku 4.5 does something I have never encountered before [0], there is a massive delta between token/sec depending on the query. Some variance including task specific is of course nothing new, but never as p
475.
by Topfi 1y ago
> [...] can you find a single criminal case in the literature where police caused Google to disgorge a complete browsing history on someone who took even modest steps not to record it (ie browsed logged out.). Can you first point to me m
476.
by Topfi 1y ago
Update, Haiku 4.5 is not just very targeted in terms of changes but also really fast. Averaging at 220token/sec is almost double most other models I'd consider comparable (though again, far to early to make a proper judgement) and
477.
by Topfi 1y ago
Hasn't that been the case throughout the industry for the last two decades now? Back when Intel was still on TikTok, the low powered laptop chips were always first, then mainstream desktop, then workstation and server roughly a year de
478.
by Topfi 1y ago
Very preliminary testing is very promising, seems far more precise in code changes over GPT-5 models in not ingesting irrelevant to the task at hand code sections for changes which tends to make GPT-5 as a coding assistant take longer than
479.
by Topfi 1y ago
> [..] single example of a court case where this happened, and my Google/ChatGPT searches are coming up completely empty. A massive amount, part of why I am both surprised and starting to feel like this discussion stems from some be
480.
by Topfi 1y ago
Yeah, politely, respectfully, no. Don't know where to start, but I want to assure you, no matter where on this planet you live, Medical Doctors are generally not at fault for high costs of care. Depending on which health care system we

More ›