reissbaker

11 ms·

61.
by reissbaker 8mo ago
Where are the polecats, though? What about the mayor's dog?
62.
by reissbaker 9mo ago
Photovoltaic production has been doubling every year. That's not a huge amount of doubling!
63.
by reissbaker 9mo ago
Counterpoint: Vim has existed for decades and does not use a bloated React rendering pipeline, and doesn't corrupt everything when it gets resized, and is much more full featured from a UI standpoint than Claude Code which is a textbox
64.
by reissbaker 9mo ago
Yes yes I'm familiar with the tweet. Nonetheless they drop frames all the time and flicker frequently. The tweet itself is ridiculous when counterpoints like Vim exist, which is much higher performance with much greater complexity. The
65.
by reissbaker 9mo ago
It's really unfortunate the term "free software" took off rather than e.g. "libre software", since it muddies discussions like this. The point of "free software" is not "you don't have to pay,&qu
66.
by reissbaker 9mo ago
Surely you mean 6fps
67.
by reissbaker 9mo ago
Generally speaking, 8xH200s will be a lot cheaper than 16xH100s, and faster too. But both should technically work.
68.
by reissbaker 9mo ago
FWIW, Ink is working on an incremental rendering system: they have a flag to enable it. It's currently pretty buggy though unfortunately. Definitely wish Anthropic would commit some resources back to the project they're built on t
69.
Synthetic.new <3 OpenCode(synthetic.new)
9 points by reissbaker 9mo ago|0 comments
70.
by reissbaker 9mo ago
If there weren't protests, Iran wouldn't have cut the internet. And besides, the BBC has videos of them: https://www.bbc.com/news/articles/cgm4y0ewe93o
71.
by reissbaker 9mo ago
Wow, that's a lot of Israeli operatives. Sounds like there are more Mossad agents living in Iran than there are Israelis living in Israel!
72.
by reissbaker 9mo ago
It worked out poorly for America — we got stuck in a long expensive war that we got basically nothing from — but for the average Iraqi? I'd much rather be an Iraqi citizen than an Iranian one, and that wouldn't have been true in
73.
by reissbaker 9mo ago
This is orthogonal to nutritious eating habits; I don't think the food pyramid should lie about nutrition due to ecological concerns. (I do think the food pyramid should be a little more concerned about saturated fat than it is, thou
74.
by reissbaker 9mo ago
That's true, although fish has a better balance of essential fatty acids than red meat. Although, oddly enough, wagyu has a (much) better fatty acids profile than other types of beef, so you can justify the occasional wallet splurge on
75.
by reissbaker 9mo ago
Broccoli has 2.8g of protein per 100g. Beef has 26g per 100g, and chicken has 27g. If you're trying to get protein, broccoli isn't going to do much, and I think it's good that the government is being honest about that. A char
76.
by reissbaker 9mo ago
Beef has ~3x more protein per gram than legumes. It is much more protein-dense than vegetables or legumes. Similarly, it's a "complete" protein, whereas most vegetables and legumes are missing necessary amino acids. The downs
77.
by reissbaker 10mo ago
The fact that virtually all new models in 2026 will have NACS tells you we don't need to regulate in 2026 that all new cars must be built with NACS. That's what's happening anyway. Once you exclude Supercharger Why would yo
78.
by reissbaker 10mo ago
The GPT-5 series is a new model, based on the o1/o3 series. It's very much inaccurate to say that it's a routing system and prompt chain built on top of 4o. 4o was not a reasoning model and reasoning prompts are very weak c
79.
by reissbaker 10mo ago
AFAIK every major car manufacturer has announced they're switching to NACS for the American market (or has already switched). I think you're underselling how standard it is. And it's already easy to get dongles for old cars!
80.
by reissbaker 10mo ago
...Curiosity satisfied, I suppose. While I disagree with some of the USAID cuts, I don't think that "not giving charity" is the same thing as "mass murder."
81.
by reissbaker 10mo ago
The U.S. already standardized on a charging port: Tesla's. https://en.wikipedia.org/wiki/North_American_Charging_Standa...
82.
by reissbaker 10mo ago
One of the greatest mass murderers in history...? I uh, am morbidly curious to hear your thought process here.
83.
by reissbaker 10mo ago
Depends on how well the speculator predicts your prompts, assuming you're using speculative decoding — weird prompts are slower, but e.g. TypeScript code diffs should be very fast. For SGLang, you also want to use a larger chunked pref
84.
by reissbaker 10mo ago
According to Wikipedia, the Yunnan mushrooms indeed have their hallucinogens broken down after cooking: https://en.wikipedia.org/wiki/Hallucinogenic_bolete_mushroom Good guess! Although, the local hospital records impl
85.
by reissbaker 10mo ago
Sadly, it's worse. We don't have one experiment that works in mice: we have dozens, if not hundreds. We've cured "Alzheimer's in mice" many times over. The treatments never work in humans, because it's not
86.
by reissbaker 10mo ago
They're not eliminating a competitor, they're (effectively) acquiring a competitor. Nvidia's GPUs are great for training, and not bad for inference, but the custom chips are better for inference and Nvidia's worried abou
87.
by reissbaker 10mo ago
This generally isn't true. Cloud vendors have to make back the cost of electricity and the cost of the GPUs. If you already bought the Mac for other purposes, also using it for LLM generation means your marginal cost is just the elec
88.
by reissbaker 10mo ago
The parser is a small part of running an LLM, and Zai's format is superior to Harmony: it avoids having the model escape JSON in most cases by using XML, so e.g. long code edits are more in-domain compared to pretraining data (where co
89.
by reissbaker 10mo ago
No, it's not Harmony; Z.ai has their own format, which they modified slightly for this release (by removing the required newlines from their previous format). You can see their tool call parsing code here: https://github.com
90.
by reissbaker 10mo ago
s/Sonnet 3.5/Sonnet 4.5 The model output also IMO look significantly more beautiful than GLM-4.6; no doubt in part helped by ample distillation data from the closed-source models. Still, not complaining, I'd much prefer a che

More ›