ImprobableTruth

7 ms·

1.
by ImprobableTruth 13d ago
Why would you accept it when the benchmark's ranking is obviously nonsense. It literally has muse spark 1.3 above 6 astra, 5.6 sol and fable 5. Anyone who has played with any of these models for any amount of time would immediately rea
2.
by ImprobableTruth 14d ago
That's the ankle. The actual knee is hidden in the feathers of the body.
3.
by ImprobableTruth 15d ago
Its "pure capabilities" are definitely worse than Fable, but I find codex has a much more pleasant style and is in comparison much more generous with its limits.
4.
by ImprobableTruth 21d ago
It's probably GLM 5.3 flash, so weaker but cheaper.
5.
by ImprobableTruth 1mo ago
These exist and are used. The issue is that because they're so much smaller, they're also much worse, so they tend to have lots of false positives while still being easy to circumvent.
6.
by ImprobableTruth 2mo ago
The fundamental issue is that for these systems bigger is essentially always better (if affordable). So if we can squash something like Kimi K3 down to run on a 'normal' system, that just incentivizes devs to increase the model si
7.
by ImprobableTruth 2mo ago
This is not what the bitter lesson is about. It's not "don't develop better methods, just scale", it's that those methods which scale best win. LeCun's work is fundamentally about devising a method that scales
8.
by ImprobableTruth 2mo ago
This is some bizarre victim inversion. The providers of closed models are the ones who are trying to use regulation to stop their open model competition, not the other way around.
9.
by ImprobableTruth 2mo ago
Now, what if you order a construction crew for building your house? What if you also hire an architect to design the plan? And an overseer who manages everything? Surely at some point you would stop saying "I built this house", ev
10.
by ImprobableTruth 3mo ago
Sure, but "10x faster, but only applies on small greenfield, throwaway projects" is a major caveat. In fact, there's a good chance this doesn't disprove the original blog post, you could be way faster on small projects b
11.
by ImprobableTruth 3mo ago
Opus 4.5/4.6 are what many people consider the first 'good' models and it's from last year/start of this year. But fine, let's say everything before gpt 5.5 was unusable crap. Then there should still be project
12.
by ImprobableTruth 3mo ago
>I just did a ~6 month project in ~2 weeks using a frontier model. Claims like this are hard for me to take seriously because 'good' models have been available since the start of the year. So, if they really 10x one's prod
13.
by ImprobableTruth 3mo ago
This is a very strange mindset. Even if you want to treat everything as sort of billable hours this doesn't really make sense because the average boss's boss's isn't paid anywhere near 144x. If a SWE spends 100 hours to
14.
by ImprobableTruth 3mo ago
Anthropic had to settle with authors because they literally pirated books! Their behavior regarding distillation is genuinely beyond parody.
15.
by ImprobableTruth 3mo ago
Except it's not just a tool. It's when a woodworker, musician or painter completely outsources their work and just marks what's wrong, sending those parts back. Yes, the final art piece might be the same, but the artist defin
16.
by ImprobableTruth 3mo ago
Shared memory of the past meant reserving a part of the memory for the GPU, which could then not be used or accessed by the CPU. If the CPU wanted to access something, it had to copy it from the GPU's section of the memory to its own.
17.
by ImprobableTruth 8mo ago
If that was the real reason, why wouldn't they just make it so that if you don't correctly use caching you use up more of your limit?
18.
by ImprobableTruth 8mo ago
The quill and ink at least communicates that it's about writing. The new one is so abstract that when I first looked at it I had no idea what I was even looking at, it certainly doesn't communicate "this is like word" to
19.
by ImprobableTruth 8mo ago
The key thing is that you teach multiplication tables in a structured, incremental manner. Yes, it's just rote memorization, but the structure makes it way easier. You don't just dump all tables on the student at once and start qu
20.
by ImprobableTruth 9mo ago
How is the raw Gemini 3 CoT accessed? Isn't it hidden?
21.
by ImprobableTruth 9mo ago
They're not making money on inference alone because they blow ungodly amounts on R&D. Otherwise it'd be a very profitable business.
22.
by ImprobableTruth 9mo ago
> These games are the starting point, but the bulk of the game is new puzzles combining mechanics from different games together Seems like the puzzles are novel, but the mechanics are not?
23.
by ImprobableTruth 9mo ago
An almost 50% price increase. Benchmarks look nice, but 50% more nice...?
24.
by ImprobableTruth 1y ago
Unfortunately not, this model is noticeably worse. I imagine horizon is either gpt 5 nano/mini.
25.
by ImprobableTruth 1y ago
Even near-perfect LLMs would benefit from the compiler optimizations that types allow. However perfect LLMs would just replace compilers and programming languages above assembly completely.
26.
by ImprobableTruth 1y ago
This is the fault of sloppy language. In Lean, _proofs_ (equivalent to functions) and _proof objects/certificates_ (values) need to be distinguished. You can't compute proofs, only proof objects. In the above quote, replace "
27.
by ImprobableTruth 1y ago
Caveat: Coercions exist in Lean, so subtypes actually can be used like the supertype, similar to other languages. This is done via essentially adding an implicit casting operation when such a usage is encountered.
28.
by ImprobableTruth 1y ago
I think the concept of a game DSL is cool, but it just feels so undercooked to me. Like, I'm a huge fan of gradual typing, especially TypeScript's, but gdscript's is just so primitive. Not even to speak of something like inte
29.
by ImprobableTruth 1y ago
What's the issue with the remix -> react-router transition? As far as I can tell it's just a branding thing.
30.
by ImprobableTruth 1y ago
I think the fact that all (good) LLM datasets are full with licensed/pirated material means we'll never really see a decent open source model under the strict definition. Open weight + open source code is really the best we'r

More ›