impure

9 ms·

61.
by impure 1y ago
The 5 cents for Nano is interesting. Maybe it will force Google to start dropping their prices again which have been slowly creeping up recently.
62.
by impure 1y ago
OpenAI and most AI companies do not train on data submitted to a paid API.
63.
by impure 1y ago
> Important numbers like this must be fair and accurate, they can’t be manipulated for political purposes Immediately attempts to manipulate them for political purposes.
64.
by impure 1y ago
PocketBase, Appwrite, Supabase... It already had its iPhone moment. I don't think it will ever be as popular as the iPhone though just because AWS has a better marketing team and self-hosted solutions tend not to be very profitable.
65.
by impure 1y ago
I find it funny that the first sentence is: I’m not sure if the whole AI thing is falling apart, but we are definitely in a phase where the hype doesn’t match what we are seeing in the real world. Also I do not think AI is falling apart and
66.
by impure 1y ago
Just use Git Submodules.
67.
by impure 1y ago
> In a move that at first went unnoticed Oh, I noticed. I've also complained how Gemini 2.0 Flash is 50% more expensive than Gemini 1.5 Flash for small requests. Also I'm sure if Google wanted to price Gemini 2.5 Flash cheaper
68.
by impure 1y ago
Pretty much all of my productivity apps has LLM integration now. My language learning app uses them to break down phrases and get detailed definitions. My RSS app generates summaries. And recently I released an email app that's like Go
69.
by impure 1y ago
I've been playing around with E4B in AI Studio and it has been giving me really great results, much better than what you'd expect from an 8B model. In fact I'm thinking of trying to install it on a VPS so I can have an altern
70.
by impure 1y ago
I suspect the difference is in the training data. Gemini is much more locked down and if it tries to repeat something from the draining data verbatim you will get a 'recitation error'.
71.
by impure 1y ago
This sounds a lot like accuracy collapse as discussed in that Apple paper. That paper clearly showed that there is some point where AI accuracy collapses extremely quickly. I suspect it has something more to do with the model producing too
72.
by impure 1y ago
I’ve been playing around with Gemma E4B and have gotten really good results. That’s a model you can run on a phone. So although prices have been going up recently I suspect they will start to fall again soon.
73.
by impure 1y ago
Mozilla should start charging a licensing fee for starting your user agent with Mozilla/ Money problems solved!
74.
by impure 1y ago
Gemini is having issues. Looks like it's time to take a break from coding and touch grass. Although my Racknerd VPS is doing fine.
75.
by impure 1y ago
I do something like this too because links in emails are insanely long. It's worse in marketing emails. So I shorten the links to save on tokens and expand them again when I get the response back from the LLM.
76.
by impure 1y ago
Interesting that they reduced the memory usage by half. This would address what is IMO the biggest problem with local LLMs: the limited number of parameters resulting in answers that are not very good. Also it's funny that they are say
77.
by impure 1y ago
Oh, so that's real sand. I thought that was camera trickery. No way someone would be crazy enough to dump real sand on a stage while someone's performing live.
78.
by impure 1y ago
I’ve had a similar problem with LLMs. They sometimes don’t properly escape quotes in JSON breaking things.
79.
by impure 1y ago
Huh, I was always told that inheritance hurt performance as it requires additional address lookups. Thats why many game engines are moving away from it. I guess it could simplify the GC but modern garbage collectors have come a long way.
80.
by impure 1y ago
I like how Nano matches Gemini 2.0 Flash's price. That will help drive down prices which will be good for my app. However I don't like how Nano behaves worse than 4o Mini in some benchmarks. Maybe it will be good enough, we'l
81.
by impure 1y ago
This is big news. Shaders have been a pain to develop for in Unity. If I can program them in Rust, leveraging Rust's tooling and ecosystem, that would be huge.
82.
by impure 2y ago
10 million token context window? Damn, looks like Gemini finally has some competition. Also I'm a little surprised this is their first Mixture of Experts model, I thought they were using that before.
83.
by impure 2y ago
The article starts off strong by making a strong argument that 'it's different this time' because of the US's trade deficit and how Americans buy too much junk and don't save enough. But then it says 'That woul
84.
by impure 2y ago
I love AI Studio. It's free to use and you can switch between models. Gemini 2.5 Pro's reasoning can be a little slow so if you don't need it you can switch to Gemini 2.0 Flash (which is capable of 100-200 tokens per second).
85.
by impure 2y ago
It’s a 27B model, I highly doubt that.
86.
by impure 2y ago
> You only opt-in to the Internet of Beefs driven by a sincere grievance if you are mook enough to want to. If you aren’t, and you haven’t fallen into the IoB by becoming vulnerable or compromised in some way, you are there either becaus
87.
by impure 2y ago
I definitely noticed I was using the apps more for entertainment than for dating. Which is why I stopped using them.
88.
by impure 2y ago
I built my own called Stratum.
89.
by impure 2y ago
Please don't.
90.
by impure 2y ago
I suspect the dichotomy of the two posts is because productivity comes naturally to many people. Others need a system. Although I do agree that a lot of people tend to overdo it and track meaningless metrics.

More ›