Hacker News: akazantsev

New comment by akazantsev in "Kimi K3.1 Pro and K3.1 Pro Highspeed"

akazantsev — Fri, 23 Oct 2026 18:33:39 +0000

Same for me, however! Check the price-per-task table. They claim to be cheaper than GLM 5.3.

Edit: or they lie... https://artificialanalysis.ai/models/kimi-k3-1-pro

New comment by akazantsev in "GLM-5.3 and the spread of advanced cyber capabilities"

akazantsev — Tue, 29 Sep 2026 22:29:34 +0000

Also, all the guys currently building huge data centers for LLMs will be extremely pissed if this results in a ban on GLM for them, and they will be locked out of this LLM pie.

New comment by akazantsev in "GLM-5.3 and the spread of advanced cyber capabilities"

akazantsev — Tue, 29 Sep 2026 22:25:25 +0000

If I wrote something like this about my competitors, I would be beaten to death by my own teammates without the CEO, project manager, or sales team even hearing about it.

There is a very dangerous thing that is very capable and available to everyone. Cranks the volume to 100% AND IT'S JUST 20% OF OUR PRICE, HURRY UP AND TRY IT.

I previously successfully used GLM 5.3 to find out how our DRM system gets bypassed, and Mythos isn't available to me...

New comment by akazantsev in "Tells of a Slop UI"

akazantsev — Sun, 27 Sep 2026 16:00:36 +0000

They do have answers in Firefox. Maybe your setup is bugging out. I clicked through them, and all of them had answers.

Is Muse free?

Muse is available for free with a usage limit. If you reach your free limit and want more, you can upgrade to a paid subscription, or wait until your free usage limit refreshes.

New comment by akazantsev in "Entering and Breaking the Avast Antivirus Sandbox Part 2"

akazantsev — Fri, 25 Sep 2026 15:40:26 +0000

Or... working viruses on Linuxes, for that matter. They always throw something about needing glibc x.x.x or higher.

New comment by akazantsev in "Markdown in /src"

akazantsev — Wed, 23 Sep 2026 10:47:11 +0000

> I think I'm with the majority of commenters here in thinking this would just wind up being clutter.

Any prompt-storing/sharing ideas floating around hit the same wall - they assume the text will be in English. It won't.

New comment by akazantsev in "Markdown in /src"

akazantsev — Wed, 23 Sep 2026 10:31:29 +0000

Oh, it will work just fine once he opens a project which he will need to modify and which is full of Chinese markdown files.

New comment by akazantsev in "Heretic removes restrictions from language models"

akazantsev — Mon, 21 Sep 2026 14:34:39 +0000

I asked GLM 5.3 to hack our DRM. I didn't even need to do anything for it to agree. Same with GLM 5.3 Flash. Make sure they have at least Python available for their task. The Flash went ahead and started reverse-engineering using PowerShell scripts and "manually" decoding bytes from its output.

New comment by akazantsev in "The Return of Sail Power: Cargo Ships Are Turning Back to the Wind"

akazantsev — Thu, 17 Sep 2026 12:21:03 +0000

> We need to start fitting nuclear reactors on cargo ships.

That's a great idea, and it's timely too. Right now, cargo ships are being attacked in many parts of the world.

New comment by akazantsev in "Big AI sets out its terms for regulatory capture"

akazantsev — Mon, 14 Sep 2026 14:46:26 +0000

It's better to buy models directly from providers that guarantee the quality of their quantization. Otherwise, model quality is a gamble. You can pay the same $10 on Novita or with any other provider.

New comment by akazantsev in "Rust is tier-1 language at Microsoft"

akazantsev — Fri, 11 Sep 2026 11:00:57 +0000

clangd alone uses 6 GB when working with Chromium’s codebase. Code indexes can be very large, even for a simple application. Out-of-the-box IDEs must index the standard library and keep it in memory.

New comment by akazantsev in "Grep beats LSP? Why coding agents ignore your fancier tools"

akazantsev — Fri, 04 Sep 2026 12:01:24 +0000

> The author is not a native speaker.

As a non-native speaker, I really hate how people use this as an excuse. LLMs are good at fixing grammatical errors, so I use them daily as spell checkers. Using LLMs for rewriting is just lazy.

Try: I'm learning English as a second language. I will provide you with a text, and you have to print the corrected text in a code block so that I can copy it. Under it, there should be two lists: one containing the errors and brief descriptions of them, and another containing tips for improving the text. Do not give your opinion about the text. Focus on grammar.

Format: [TEXT] C: [CONTEXT]

New comment by akazantsev in "LLMs: Intelligence vs. Cost"

akazantsev — Wed, 02 Sep 2026 21:31:35 +0000

> This is fine in most cases, but for open-weights models it can be a lot more expensive than what the exact same model can be rented for from third-party API providers.

Filter by quantization, and most providers will have the same price. There is some "base" price even for open-weight models. Anything cheaper means some tricks on the provider's side.

New comment by akazantsev in "Vivaldi for Android now supports some Chrome extensions"

akazantsev — Tue, 01 Sep 2026 21:24:32 +0000

Hopefully, they will do just what Brave did - embed an ad blocker compatible with uBlock lists right into the browser. This way, Manifest V3 might blow up in Google's face if a sufficient number of browsers come with ad blockers out of the box.

That being said, they are not talking about Manifest V3 extensions. It's just that they can't test all the extensions and will have to work out the issues as they implement the extension API for mobile.

New comment by akazantsev in "New Mac Studio with M5 Max and M5 Ultra"

akazantsev — Thu, 27 Aug 2026 10:09:40 +0000

It's not open source.

https://en.wikipedia.org/wiki/LM_Studio

New comment by akazantsev in "Better Batteries"

akazantsev — Fri, 21 Aug 2026 12:26:02 +0000

> Java also used to have a big standard library, with many useful things including GUI programming which has been neglected a little bit.

Swing was replaced with JavaFX, then JavaFX was removed from JDKs. Long live Swing! But yes, it's extremely nice to have a GUI library out of the box.

New comment by akazantsev in "Canonical backs new project to translate large C codebases into safe Rust"

akazantsev — Thu, 20 Aug 2026 15:45:00 +0000

Demanding others to pick up Rust "for fun" is not fun at all.

New comment by akazantsev in "Canonical Backs New Project to Translate Large C Codebases into Safe Rust"

akazantsev — Thu, 20 Aug 2026 14:32:05 +0000

> capable engineers can pick up any language

Sure, given that they are getting paid for it. How many people will be interested in learning a programming language with insignificant job availability, in addition to contributing for free?

New comment by akazantsev in "Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows"

akazantsev — Mon, 10 Aug 2026 13:45:31 +0000

> The submission’s title (open weights 30B local coding model) is luckily wrong: This is meant to be a general agentic model.

https://xcancel.com/alexandr_wang/status/2086756152034066792

It's correct. See the OpenCode demo. Generic models are good enough for coding without necessarily being designed specifically for coding.

New comment by akazantsev in "Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users"

akazantsev — Fri, 07 Aug 2026 12:03:49 +0000

Those are just the default settings. Here, you can force it not to use emojis and to be much less verbose: https://chatgpt.com/?no_universal_links=1#settings/Personali...