maldie

9 ms·

1.
by maldie 1mo ago
Even with "low" thinking effort and Q4 it makes reasonable pelican, only 4k tokens and half a minute on 4090: https://imgur.com/a/yd5rNCp
2.
by maldie 3mo ago
llama.cpp to get 115 tok/s on RTX 4090 with Qwen3.6-27B. For example in Windows the latest CUDA variant llama-b9678-bin-win-cuda-13.3-x64.zip and Unsloth UD-Q4_K_XL MTP gguf: llama-server.exe --host 0.0.0.0 --alias "Qwen3.6-27B-MT
3.
by maldie 11mo ago
Yes, disassembling and tinkering with microwave is almost the only easily available thing I would not recommend at all because it seems so ordinary but actual danger is way too high and hidden. Even if you know exactly the dangers, it takes
4.
by maldie 1y ago
For sure seems they are likely deliberately riding on the fame of the movie. I too instantly thought it is some kind of Flow movie animation collaboration similarily like Flow is represented in Blender 4.4 splash screen or is even their mas
5.
by maldie 2y ago
Interesting what power do microwaves in US deliver. In EU the regular micros are about 900W range give or take 200W (that is 1500W from wall because microwaves are not efficient machines) but kettles have usually range 2200..3000. So it is
6.
by maldie 2y ago
This is complete insanity that vertical task bar is still not implemented, pretty much what OS has to do - manage memory, manage processes/threads, manage windows to enable smooth UX with multiple programs. It more and more fails on th
7.
by maldie 4y ago
Agreed, expectations are everything or at least very important to keep sanity in this area. Reality is that big majority of startup-s do fail to "make it" 90% of time to even be self sustaining. Financially meaning that you pay fr