srigi

Joined 11 years ago

Karma
420
Submissions
88
Member since
Jul 14
srigi.at.hn meet.hn/city/cz-Brno

Socials: - github.com/srigi - x.com/srigi

Interests: Freelancing, Hardware, Programming, Technology ---

View on Hacker News

Recent submissions

Comment · 20 days ago
Comment · 2 months ago
People when OW LLM looks good in benchmarks: benchmaxxxed

People when OW LLM looks mediocre in benchmarks: disappointed

Comment · 4 months ago
Comment · 4 months ago
In (post)communist countries there is a common name - ISOSTAT switch.
Comment · 5 months ago
Not yet, but soon… Bun
Comment · 6 months ago
You could keep multimodal projector (understanding of audio, images & PDFs) in system RAM with `--no-mmproj-offload` in llama.cpp. Of course, then it is not accelerated with GPU, but you save its VRAM.
Comment · 6 months ago
Comment · 7 months ago
The official SW on MacOS is somewhat usable if:

- you disable all communication with a firewall (so it doesn't autoupdate)

- `sudo pkill -9 LogiPluginService && sudo rm -rf /Applications/Utilities/LogiPluginService.app` (so it does not eat resources and don't run a useless service in the bg)

Comment · 9 months ago
"I'm sorry, my knowledge cuttoff is 1875"
Comment · 1 year ago
Start with the YT series on neural nets and LLMs from 3blue1brown
Comment · 1 year ago
Comment · 1 year ago
Yeah, I remember that option - basic tier of DTU DB with 250GB of storage - free for one year, then continue for $15/m.

When the client brought some 3rd party expert and he advised rewriting to MySQL, I quickly did the math and it was like $60/m, without a free year.

We continued with DTU MSSQL with Prisma ORM and never regreted.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton