Hacker News: om8

New comment by om8 in "TurboQuant: Redefining AI efficiency with extreme compression"

om8 — Wed, 25 Mar 2026 09:58:53 +0000

These are very different media types with very different goals.

New comment by om8 in "Claude Opus 4.6"

om8 — Thu, 05 Feb 2026 18:23:41 +0000

Is there a way to disable it? Sometimes I value agent not having knowledge that it needs to cut corners

New comment by om8 in "Linux Kernel Rust Code Sees Its First CVE Vulnerability"

om8 — Wed, 17 Dec 2025 18:40:20 +0000

https://steveklabnik.com/writing/does-unsafe-undermine-rusts...

New comment by om8 in "You are the scariest monster in the woods"

om8 — Wed, 15 Oct 2025 16:14:32 +0000

... Yet.

New comment by om8 in "Uv overtakes pip in CI"

om8 — Tue, 14 Oct 2025 05:03:25 +0000

FYI you can run just `uvx pdm`

New comment by om8 in "Mass phishing emails pretending to be Y Combinator right now"

om8 — Tue, 23 Sep 2025 23:07:05 +0000

Also got it, found this thread by googling "ycombiinator"

New comment by om8 in "GitHub Y Combinator phishing attack"

om8 — Tue, 23 Sep 2025 23:06:48 +0000

Also got it, found this thread by googling "ycombiinator"

New comment by om8 in "RustGPT: A pure-Rust transformer LLM built from scratch"

om8 — Mon, 15 Sep 2025 21:50:58 +0000

Have a similar project. Also written in rust, runs in a browser using web assembly

In-browser demo: https://galqiwi.github.io/aqlm-rs

Source code: https://github.com/galqiwi/demo-aqlm-rs

New comment by om8 in "Website is hosted on a disposable vape"

om8 — Sun, 14 Sep 2025 23:22:00 +0000

https://web.archive.org/web/20250914222638/http://ewaste.fka...

New comment by om8 in "Anna's Archive: An Update from the Team"

om8 — Tue, 19 Aug 2025 06:41:05 +0000

Oh, ok. Thanks, I agree

New comment by om8 in "Anna's Archive: An Update from the Team"

om8 — Mon, 18 Aug 2025 18:59:51 +0000

Why? They acquired books, that’s what they do

New comment by om8 in "Why are there so many rationalist cults?"

om8 — Tue, 12 Aug 2025 20:43:17 +0000

Good rationalism includes empiricism though

New comment by om8 in "Ollama and gguf"

om8 — Mon, 11 Aug 2025 23:22:52 +0000

llama.cpp is a mess and ollama is right to move on from it

New comment by om8 in "OpenAI's new GPT-5 models announced early by GitHub"

om8 — Thu, 07 Aug 2025 10:11:39 +0000

Of course it is. GPT-5 is one of the most anticipated things in AI right now. To live up to the hype, it needs to be a reasoning model.

New comment by om8 in "Ollama Turbo"

om8 — Tue, 05 Aug 2025 23:36:55 +0000

It’s unfortunate that llama.cpp’s code is a mess. It’s impossible to make any meaningful contributions to it.

New comment by om8 in "Show HN: WebGPU enables local LLM in the browser – demo site with AI chat"

om8 — Sat, 02 Aug 2025 17:21:17 +0000

Sure, but integrated graphics usually lacks vram for LLM inference.

New comment by om8 in "WebGPU enables local LLM in the browser. Demo site with AI chat"

om8 — Sat, 02 Aug 2025 16:48:44 +0000

To have a gpu inference, you need a gpu. I have a demo that runs 8B llama on any computer with 4 gigs of ram

https://galqiwi.github.io/aqlm-rs/about.html

New comment by om8 in "Show HN: I made a website that makes you cry"

om8 — Fri, 01 Aug 2025 08:36:32 +0000

I'm currently in Russia trying to get a US visa for my CS PhD. Because I do CS, I got into a thing called administrative processing. For 95% of people, it takes days -- weeks, tops. Because of the colour of my passport, it has already lasted for 3 months. I know people who are waiting for 2 years to pass it.

Why do you think I shouldn't have access to this website in Russia?

New comment by om8 in "Show HN: I made a website that makes you cry"

om8 — Fri, 01 Aug 2025 08:12:00 +0000

I believe he knows

New comment by om8 in "Uv: Running a script with dependencies"

om8 — Tue, 22 Jul 2025 04:56:03 +0000

What would you do if some of the deps started to have conflicts in them? Also, what are your plans for migration when you'll need to move from one os version to another?

Implicit solutions like yours have lower cost of entrance, but larger cost of support. uv python scripts just work if you set them up once