Hacker News: Oras

New comment by Oras in "France confirms data breach at government agency that manages citizens' IDs"

Oras — Thu, 23 Apr 2026 16:14:24 +0000

are govs required to comply with GDPR and data breaches laws?

New comment by Oras in "Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model"

Oras — Wed, 22 Apr 2026 16:42:06 +0000

If these models reach quality of Opus 4.5, then DGX could be a good alternative for serious dev teams to run local models. It is not that expensive and has short time to make ROI

New comment by Oras in "Show HN submissions tripled and now mostly have the same vibe-coded look"

Oras — Wed, 22 Apr 2026 15:09:06 +0000

I agree. The problem is the noise ratio, not how the platform was implemented.

New comment by Oras in "ChatGPT Images 2.0"

Oras — Tue, 21 Apr 2026 22:40:55 +0000

My test for image models is asking it to create an image showing chess openings. Both this model and Banana pro are so bad at it.

While the image looks nice, the actual details are always wrong, such as showing pawns in wrong locations, missing pawns, .. etc.

Try it yourself with this prompt: Create a poster to show opening game for Queen's Gambit to teach kids to play chess.

New comment by Oras in "Claude Code to be removed from Anthropic's Pro plan?"

Oras — Tue, 21 Apr 2026 22:19:55 +0000

Incredible, powerful, but I couldn't believe how fast I hit the limits compared to how it was with Opus 4.6. They removed Opus 4.6 completely from CC. I would prefer it with the previous limits.

That's not how you keep your customers. None of these agents have a moat, I moved away from Cursor when they started doing what Anthropic is doing now, and never went back even when I was a paying customer since the start.

New comment by Oras in "Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving"

Oras — Mon, 20 Apr 2026 15:28:28 +0000

Not just aesthetics though, from time to time I implement the same feature with CC and Codex just to compare results, and I yet to find Codex making better decisions or even the completeness of the feature.

For more complicated stuff, like queries or data comparison, Codex seems always behind for me.

New comment by Oras in "Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving"

Oras — Mon, 20 Apr 2026 15:24:39 +0000

cheaper is quite subjective, I just went to their pricing page [0] and cost saving compared to performance does not sell it well (again, personal opinion).

CC has a limited capacity for Opus, but fairly good for Sonnet. For Codex, never had issues about hitting my limits and I'm only a pro user.

https://z.ai/subscribe

New comment by Oras in "Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving"

Oras — Mon, 20 Apr 2026 14:42:41 +0000

I find it odd that none of OpenAI models was used in comparison, but used Z GLM 5.1. Is Z (GLM 5.1) really that good? It is crushing Opus 4.5 in these benchmarks, if that is true, I would have expected to read many articles on HN on how people flocked CC and Codex to use it.

New comment by Oras in "GitHub's Fake Star Economy"

Oras — Mon, 20 Apr 2026 09:44:00 +0000

Would be nice to see the ratio of OpenClaw stars

New comment by Oras in "Claude Design"

Oras — Fri, 17 Apr 2026 15:44:35 +0000

Data suggest different outcomes, there was always a way to standardise interfaces, from Twitter bootstrap, all the way to shadcn.

Not everyone is looking for unique design, 70% of the web is still using Wordpress. I would say majority prefer familiarity and appreciate uniqueness.

New comment by Oras in "Scientists invented a fake disease. AI told people it was real"

Oras — Fri, 10 Apr 2026 12:29:12 +0000

you're confusing LLMs with humans

New comment by Oras in "Scientists invented a fake disease. AI told people it was real"

Oras — Fri, 10 Apr 2026 10:21:04 +0000

This would work on people too, you can see daily fake info/text/videos and many people believing in them.

LLMs do not think, why this is still hard to understand? They just spit out whatever data they analyse and trained on.

I feel this kind of articles is aimed at people who hate AI and just want to be conformable within their own bias.

New comment by Oras in "I still prefer MCP over skills"

Oras — Fri, 10 Apr 2026 09:30:40 +0000

> majority of users on this planet don't use AI agents like that

Source?

New comment by Oras in "CLI proxy that reduces LLM token consumption by 60-90% on common dev commands"

Oras — Fri, 10 Apr 2026 08:34:21 +0000

It looks so promising, but the first thing came to my mind is these models are mostly trained on the default cli output, would compressing it mess with the output of these models?

New comment by Oras in "Claude Mythos Preview: Available in Private Preview on Vertex AI"

Oras — Wed, 08 Apr 2026 11:23:36 +0000

> available in Private Preview to a select group of Google Cloud customers, as part of Project Glasswing

New comment by Oras in "Škoda DuoBell: A bicycle bell that penetrates noise-cancelling headphones"

Oras — Wed, 08 Apr 2026 09:48:42 +0000

Fines. No one should cross roads/paths randomly, with or without headphones.

One large fine, and people will learn.

New comment by Oras in "Škoda DuoBell: A bicycle bell that penetrates noise-cancelling headphones"

Oras — Wed, 08 Apr 2026 09:33:43 +0000

Over engineering in real life, solving lack of common sense by introducing a solution where the cyclist is paying.

I think the solution is nice for sure, but solving the wrong problem.

New comment by Oras in "Employers use your personal data to figure out the lowest salary you'll accept"

Oras — Mon, 06 Apr 2026 07:37:07 +0000

They didn’t ask for any of that when harvesting personal financial data, how lovely

New comment by Oras in "Scientists found a protein that drives brain aging – and how to stop it"

Oras — Mon, 06 Apr 2026 07:26:08 +0000

Also focusing on memory and metabolism, I thought it’s about aging in general such as muscles and mobility

New comment by Oras in "Ask HN: Anyone started a solo business in the last 6 months and made it work?"

Oras — Sat, 04 Apr 2026 18:23:26 +0000

The timeframe of 6 months is kind of a fantasy. Building a business or successful product takes way more than that. It’s a marathon not a sprint.

The outliers that you might see are one of the following:

- Someone with large following/authority built over many years.

- Burning cash in ads, mainly VC funded, not solo.

I’m writing this hoping it will help someone to realise reality and avoid falling for online traps that building a business is easy or can be done in months.