k__

4 ms·

31.
by k__ 2mo ago
I didn't get the impression that anyone competed with the old prices before.
32.
by k__ 2mo ago
I get like 80.
33.
by k__ 2mo ago
I wouldn't exactly call it snappy, but faster than Pro, yes.
34.
by k__ 2mo ago
I tried the previous Pro model and in the end it was 50% more expensive than the previous Flash. Wasn't worth it.
35.
by k__ 2mo ago
Around 5 percentage points better. (E.g., 87% instead of 82%)
36.
by k__ 2mo ago
Had the same impression about TypeScript and Rust. Not as fun to write as Python and Nim, but I don't have to write it.
37.
by k__ 2mo ago
Care to elaborate?
38.
by k__ 2mo ago
"A local Qwen3.5 retrieval model attends over the indexed memory" Hrm.
39.
by k__ 2mo ago
Seems like this only helps with parallel workloads.
40.
by k__ 2mo ago
Location: Germany Remote: Yes Willing to relocate: No Roles: Technical Writer, Software Engineer Homepage: https://kay.is LinkedIn: https://www.linkedin.com/in/kay-plößer
41.
by k__ 2mo ago
Prompt an image or video generator without knowledge in photography or art skills and your results will look sloppy.
42.
by k__ 2mo ago
I did a bit of research on that in my last job and got the impression that encoder models might help. They are well suited to check input for rule violations and are much cheaper to train and run than decoder models. The downside is, they c
43.
by k__ 2mo ago
Yeah, I'd assume it's possible to extract all languages as steering vectors from a model and then substract the ones you don't need from its weights. However, that would just change the weights values and not their dimensions
44.
by k__ 2mo ago
I think, the main issue with German rules is that we haven't embraced digital technology 100%. All these rules would be way less cumbersome if they didn't come with a bunch of literal paperwork.
45.
by k__ 2mo ago
The ratio of good books to slop (AI or not) was far to bad for humans to reasonably filter 20 years ago.
46.
by k__ 2mo ago
Haha, and I almost felt bad after seeing this chart yesterday.
47.
by k__ 2mo ago
Half OT: Why do the cache hit rates seem to vary so much between harnesses? I use pi, which is very minimalist, and I get a hit rate of ~99%. Paying like $1 a day for Flash. Yet, the hit rate mentioned on OpenRouter is only ~79%.
48.
by k__ 2mo ago
On OpenRouter it's 93 TPS.
49.
by k__ 2mo ago
I'm using pi and my caching is ~99%.
50.
by k__ 2mo ago
Yeah, it needs quite some hand holding. I didn't do much agent coding and had a mix experience. 1. It would build something that was in the spirit of what I wanted, but unusable in practice. 2. It would build something quite useful, bu
51.
by k__ 2mo ago
I'd take more throughput while everything else stays the same.
52.
by k__ 2mo ago
While I tend to clear my session after every task, I feel less stressed when my context is as 10% than when it's at 30%
53.
by k__ 2mo ago
They remove potentially irrelevant details.
54.
by k__ 2mo ago
As I understand it, they would have to train a whole knew model to hard cap it's context to different lengths. That would be cheaper to train and had cheaper inf, but still a huge investment. So I'd guess it's API level.
55.
by k__ 2mo ago
Right
56.
by k__ 2mo ago
Yeah, I think the Effect team tried building a compiler for once (TS++ or something) but they abandoned it, as it was too much work.
57.
by k__ 3mo ago
Do crosswords count?
58.
by k__ 3mo ago
My 2 weeks with DeepSeek V4: Pro is ~50% more expensive than Flash. Both need babysitting. Plan, split in small tasks, give it docs, types, tests, linter, best practice examples, etc. Always start a new session when starting a task. Do regu
59.
by k__ 3mo ago
Thanks! What is the parento frontier?
60.
by k__ 3mo ago
Half-OT: can anyone recommend a LLM cost calculator that's up to date?

More ›