betimsl

15 ms·

1.
by betimsl 3mo ago
The problem with these is the tool calling. From my experiments qwen agent almost always fails with tool calling and porting the correct config is quite tedious. Rio3.5 with Qwen compatible tool calling, we need that :)
2.
by betimsl 3mo ago
what a legend!
3.
by betimsl 4mo ago
PocketBase is awesome and this thing looks awesome. One question, I could not understand if a mail provider is needed or it handles on its own?
4.
by betimsl 10mo ago
Could you please make the book into a PDF?
5.
by betimsl 11mo ago
Who wrote the algorithm?
6.
by betimsl 11mo ago
More important that if it is a good programmer or not, is to be able to spot if the guy can get the job DONE.
7.
by betimsl 1y ago
The story is that Ken and Rob were at a diner when Ken gave structure to it and wrote the initial encode/decode functions on napkins. UTF-8 is so simple yet it required a complex mind to do it.
8.
by betimsl 1y ago
Right, but Intel is in no position to do that. So, they could play and put 32GB VRAM and also, why not produce one with 64GB just for kicks?
9.
by betimsl 1y ago
Dream phone: Underclocked Samsung25 internals, iPhone SE design/dimensions, at least 90hz OLED panel, 1 back decent camera, 1 front too. 3A battery. Preferably an extenal SD card slot.
10.
by betimsl 1y ago
Awesome.
11.
by betimsl 1y ago
I think you could easily give the UI a refresh, modernize it a bit :)
12.
by betimsl 1y ago
It's simply not Apple's responsibility. Person's data, their responsibility. Any other way and people lose trust in Apple which BTW they protect it by any means. Trust. Instead of wanting something impossible, accept the real
13.
by betimsl 1y ago
Apple’s encryption, is designed with end-to-end encryption for many types of data. Some facts: Only the user's devices hold the keys to decrypt the data. Apple cannot decrypt it, even if served a subpoena. Apple chose pr
14.
by betimsl 1y ago
Uhoh! -- The virus just spread a litl more!
15.
by betimsl 1y ago
Check out NVIDIAs latest releases. Paying for tokens is going to be a history in about 6 months. You run the model on your laptop. Maybe you're right about the confusion...but given the velocity, that's going to be fixed also. All
16.
by betimsl 1y ago
They probably screw the patients too.
17.
by betimsl 1y ago
Why separate? It's not like there's limited capacity...you can have an assistant knowing all the languages at once.
18.
by betimsl 1y ago
You have a typo in the snippet on README.md: SchlüsselWert should be Schlüsselwert.
19.
by betimsl 2y ago
> [...] This is to the point where is starts to become hard for you to work without one. Why would one work without one?
20.
by betimsl 2y ago
> Why? Huge interest in the field. Schools unable to teach the craft properly.
21.
by betimsl 2y ago
Glad that I'm not the only one doing just that.
22.
by betimsl 2y ago
Any chance open sourcing it? :)
23.
by betimsl 2y ago
Material tolerances.
24.
by betimsl 2y ago
Running a 7b coder in laptops with 4060 is possible and with very good results. Orra looks like a very good tool to be integrated with any IDE. Take a look at this: https://github.com/huggingface/llm.nvim -- it has a b
25.
by betimsl 2y ago
I mean, local LLM, for example: can it be run with llamafile instead of cloud AI?
26.
by betimsl 2y ago
I don't see local backend in documentation. Is it possible?
27.
by betimsl 2y ago
Reads nicely from right to left.
28.
by betimsl 2y ago
how much is the cost of a kW in Germany?
29.
by betimsl 2y ago
Equivalent to a beginner betting $500 on red. Let's see what happens.
30.
by betimsl 2y ago
It just makes everything perfect.

More ›