You are free to use AI to write Zig code just not use AI to write code for Zig. In fact, many people already do build software with zig written by AI. Check out rockorager's github.
Schools like MIT pay PhD students barely above or sometimes below the poverty level of that particular state as monthly stipend. Yeah, research funding got slashed but if they had the will they could have come up with the money for that 20%.
OpenAI is very good in terms of not having as much outages as Anthropic, but almost all products except Codex and the pro model is unimpressive, anthropic has the opposite situation.
There's no way I am using such an important piece of life as an IDE from Google just because I know they are going to kill it within 3 years, if it survives that much. Probably will die with the Windsurf guy jumping ship again.
Still use pinboard which seems more than enough for me. But I would happily pay for a pinboard that has better search and more responsive customer service.
I still haven't got access to GPT-5 (plus user in US), and I am not really super looking forward to it given I would lose access to o3. o3 is a great reasoning and planning model (better than Claude Opus in planning IMO and cheaper) that I use in the UI as well as through API. I don't think OpenAI should force users to an advanced model if there is not a noticeable difference in capability. But I guess it saves them money? Someone posted on X how giving access to only GPT-5 and GPT-5 thinking reduces a plus user's overall weekly request rate.
It works fine I would say. In the absence of a bigger screen with proportional DPI (kindle scribe is 300 which is one of the highest in terms of screen size), kindle scribe is one of the better options still IMO.
I wish they would do a bigger size kindle scribe. I read pdfs all day on my scribe, and often I wish the screen was bigger so the font size would be large.
A lot of this can be explained by just supply and demand. In central New Jersey, the number of people who need to get a home is always increasing, but number of houses being built is extremely low, and even those houses start from 900k-1M.
To be honest, when I say it has significantly worsened, I am comparing to the time when GPT-4 just came out. It really felt like we were on the verge of 'AGI'. In 3 hours, I coded up a complex piece of web app with chatgpt which completely remembered what we have been doing the whole time. So, it's sad that they have decided against the public having access to such strong models (and I do think it's intentional, not some side-effect of safety alignments though that might have contributed to the decision).
My use of ChatGPT has just organically gone down 90%. It's unable to do any sort of task of non-trivial complexity e.g. complex coding tasks, writing complex prose that conforms precisely to what's been asked etc. Also I hate the fact that it has to answer everything in bullet points, even when it's not needed, clearly rlhf-ed. At this point, my question types have become what you would ask a tool like perplexity.
This has exactly been my experience for at least the last 3 months. At this point, I am thinking if paying that 20 bucks is even worth anymore which is a shame because when gpt-4 first came out, it was remembering everything in a long conversation and self-correcting itself based on modifications.
The main problem I have faced with the whisper model (large) is when there is silence or a sizable gap without audio, it hallucinates and just puts out some random gibberish repeatedly until the transcription ends. How does this app handle this?
It's not confirmation bias. Proof is I did the same, built two different webapps with languages I had little experience in, both before and after chatGPT got worse. In the latter case, I struggled much to remind it again and again of the context we are working in and had to give the current code state repeatedly as input for it to output modified code. It clearly can't remember the context and produces subpar code in a non-trivial project.
The library might be great. But from the examples shown, there is zero reason to use it as all of these can be done easily without giving your data to OpenAI.
On many contentious issues in US, Wikipedia's articles are most of the time left leaning. This is not surprising given most of its editors are left leaning. My point is, I think we are passed the stage of taking seriously anytime these big institutions batter each other with supposed moral superiority. Having said that, Musk bending to Governments like China, where he has large financial interests was a concern from even before he took over Twitter.
NPR is indeed a propaganda machine. But only for one party, and its woke ideology. It doesn't even have to receive federal funding to acknowledge that. Now the technical question is what label do you put on a media platform that's captured by an ideology favorable to a single party which comes to power every few years?
>'We release all our models to the research community'
Where release means you fill out a form and wait indefinitely. Also no use for commercial purposes - which means 95% of users are out - certainly doesn't democratize LLMs.
I bought a framework laptop 1 year ago, and trust me when I say I tried everything out there to improve the battery life on Ubuntu. It was still so bad I was always worried about the battery draining out, not to mention, just closing the lid doesn't work. Eventually switched to an m2 mac air. I barely have to think about the battery now, so I can focus on the actual work, and when work finishes I just close the lid.
The medium article you posted does analysis on OpenAI's old embeddings. OpenAI has a single new embedding model [0] that replaces all the old models and it's also super cheap ($0.0004 / 1K tokens).
Whenever someone says 'religions' in the western hemisphere, it usually means Christianity. E.g. Islam fundamentally will not have a crisis even if alien life is discovered because it doesn't disprove the existence and oneness of a Creator which is Islam's strict monotheism denotes.