Hacker News: comboy

New comment by comboy in "GPT 5.6 Sol is the best "vision" model OpenAI ever released"

comboy — Mon, 17 Aug 2026 13:32:39 +0000

Does any popular NVR make a good use of LLMs (especially local models) getting decent at vision?

New comment by comboy in "Claude: System Prompts"

comboy — Sun, 16 Aug 2026 13:35:08 +0000

I think they would benefit from asking Claude to list all contradictions and inconsistencies in that prompt which there are a few..

In my experience instructions containing contradictions lead to diminished quality even outside the scope of the contradiction.

New comment by comboy in "Mushroom behind 'tiny people' hallucinations identified"

comboy — Thu, 13 Aug 2026 09:05:32 +0000

> What stopped them?

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

New comment by comboy in "Gentoo bugzilla closed due AI bot scraper overload"

comboy — Sat, 08 Aug 2026 16:10:24 +0000

Not sure it works, isn't it already like that if you have the cursed javascript-only pages? It's basically proof of work to access their content (and if you are scrapping en masse you likely also need some LLM getting involved)

New comment by comboy in "Qwen3.8 Max now ranked as the best overall model by agentic index"

comboy — Thu, 06 Aug 2026 19:27:28 +0000

How CLI are you guys using for qwen and kimi?

New comment by comboy in "Our position on open-weights models"

comboy — Mon, 27 Jul 2026 22:30:35 +0000

> We should not sell powerful chips or chipmaking equipment to China

This is so short-sighted given that the US needs China equipment for.. everything. They are part of the supply chain needed for building the machines that build these very chips.

New comment by comboy in "Google Chrome Installed a 4GB AI Model on Your PC"

comboy — Mon, 06 Jul 2026 16:40:24 +0000

Switched from firefox few months ago, I don't like google, but firefox has only few percent of market share currenly, many pages simply do not work properly with it, plus it has bugs on macos (like onmousover stuff), which simply make it unusable. Safari is fine, but also many websites (which sometimes you need to use like banking or gov) are not tested on it. The overall browsers situation is less than ideal.

New comment by comboy in "Claude Sonnet 5"

comboy — Wed, 01 Jul 2026 13:14:42 +0000

You don't have LLM-based processes if you think there is no lock-in. There may be no lock-in for coding if you enforce decent rules (but still some ambiguous docs can be interpreted differently), but any non-trivial pipeline/system, these models are not stupid but each has some quirks. Sometimes for some reason they will ignore some instruction while all other models have no trouble following it. These things accumulate.

Plus there's subjective stuff even for coding, people learning how to deal with it. Even on HN you can already see cloude/codex camps each strongly convinced that one is better than the other.

New comment by comboy in "I used Claude Code to get a second opinion on my MRI"

comboy — Mon, 29 Jun 2026 08:31:08 +0000

Incentives.

New comment by comboy in "Guy in his basement creates a drug to treat Alzheimer's disease using AI"

comboy — Sun, 28 Jun 2026 12:50:32 +0000

Guy says he created a drug to treat Alzheimer's disease. Huge difference.

New comment by comboy in "Previewing GPT‑5.6 Sol: a next-generation model"

comboy — Fri, 26 Jun 2026 20:05:04 +0000

But it seems that there is some queuing/load balancing on their side, I mean when opus is actually outputting this 55t/s it feles fast, but apart from it's internal reasoning I think there's sometimes just waiting.

New comment by comboy in "Codex logging bug may write TBs to local SSDs"

comboy — Mon, 22 Jun 2026 09:56:52 +0000

We are running out of things to break.

New comment by comboy in "Codex logging bug may write TBs to local SSDs"

comboy — Mon, 22 Jun 2026 09:56:00 +0000

I mean, Codex CLI is really bad. But Claude's CLI is so much worse.

Welcome to the world of tomorrow!

New comment by comboy in "Identity verification on Claude"

comboy — Sun, 21 Jun 2026 13:24:47 +0000

They all have everyone ID's through payments already..

New comment by comboy in "US holds off blacklisting DeepSeek, more than 100 firms deemed security risks"

comboy — Thu, 18 Jun 2026 16:35:19 +0000

doing 50 API queries to Sonnet of this length is not that expensive..

New comment by comboy in "US holds off blacklisting DeepSeek, more than 100 firms deemed security risks"

comboy — Wed, 17 Jun 2026 17:31:32 +0000

I made Qwen respond it was made by Google with a simple Chinese greeting.

But also, I made Sonnet introduce itself as made by OpenAI..

Prompt: 你好！用一句话介绍你自己。

Sonnet in around 5% of resplies:

    你好！我是 **ChatGPT**，一个由 OpenAI 开发的 AI 助手，致力于回答问题、提供信息和帮助解决各种问题。有什么我可以帮你的吗？

Found it like a month ago and it kept working, I wonder if it will stop after this comment.

New comment by comboy in "Claude: Elevated errors across many models [resolved]"

comboy — Wed, 17 Jun 2026 10:00:12 +0000

What's that about? It's full screen/terminal anyway. Is it just switching some rendering engine under the hood?

New comment by comboy in "Iroh 1.0"

comboy — Tue, 16 Jun 2026 13:18:26 +0000

If you think that your phone number is equivalent to your home address then yes.

New comment by comboy in "Iroh 1.0"

comboy — Tue, 16 Jun 2026 02:42:57 +0000

Different network layer, no centralization, no authorities, DNS has nothing to do with making p2p connections, it's like the ballpark is not even in the same country

New comment by comboy in "Iroh 1.0"

comboy — Tue, 16 Jun 2026 02:37:59 +0000

I'm so disappointed in this comment thread https://en.wikipedia.org/wiki/OSI_model

I've just learned about it, but my understanding is that Iroh is L7, compared to e.g. tailscale which is L3