Hacker News: ZeroCool2u

New comment by ZeroCool2u in "GPT-5.5"

ZeroCool2u — Thu, 23 Apr 2026 18:07:38 +0000

Benchmarks are favorable enough they're comparing to non-OpenAI models again. Interesting that tokens/second is similar to 5.4. Maybe there's some genuine innovation beyond bigger model better this time?

New comment by ZeroCool2u in "Measuring Claude 4.7's tokenizer costs"

ZeroCool2u — Fri, 17 Apr 2026 17:05:52 +0000

Whenever we get the locally runnable 4k models things are going to get really awkward for the big 3 labs. Well at least Google will still have their ad revenue I guess.

New comment by ZeroCool2u in "Decisions that eroded trust in Azure – by a former Azure Core engineer"

ZeroCool2u — Thu, 02 Apr 2026 22:00:32 +0000

In my experience Azure is full of consistency issues and race conditions. It's enough of an issue that I was talking about new OpenAI models becoming available via Bedrock on AWS and how convenient that was since I wouldn't have to deal with Azure and my colleague in enterprise architecture went on an unprompted rant about these exact issues. It's not the first time something like this has happened and I've experienced these issues first hand, so yes. I'd say reliability is a critical issue for Azure and it hasn't gotten better each time I've gone back to check.

New comment by ZeroCool2u in "Quad9 Enables DNS over HTTP/3 and DNS over QUIC"

ZeroCool2u — Thu, 02 Apr 2026 19:05:04 +0000

Does quad9 have a resolver that includes ad blocking?

New comment by ZeroCool2u in "Claude Code users hitting usage limits 'way faster than expected'"

ZeroCool2u — Tue, 31 Mar 2026 13:59:08 +0000

Yes, exactly my point.

New comment by ZeroCool2u in "Claude Code users hitting usage limits 'way faster than expected'"

ZeroCool2u — Tue, 31 Mar 2026 13:48:19 +0000

I'm finishing my annual paid Pro Gemini plan, so I'm on the free plan for Claude and I asked one (1) single question, which admittedly was about a research plan, using the Sonnet 4.6 Extended thinking model and instantly hit my limit until 2 PM (it was around 8 or 9 AM).

Just a shockingly constrained service tier right now.

New comment by ZeroCool2u in "Project Nomad – Knowledge That Never Goes Offline"

ZeroCool2u — Sun, 22 Mar 2026 16:37:27 +0000

See I really want this in a simpler format. Like a single file embedded database on my filesystem that I can point a single/or few tools at for my model to use when it needs.

New comment by ZeroCool2u in "Atlassian says it had right to fire engineer for suggesting CEO is 'rich jerk'"

ZeroCool2u — Sun, 22 Mar 2026 16:02:07 +0000

Gift link then: https://www.bloomberg.com/news/articles/2026-03-16/atlassian...

New comment by ZeroCool2u in "Aliens.gov ~ domain registered 17MAR2026"

ZeroCool2u — Wed, 18 Mar 2026 14:27:48 +0000

More like over $5.00 per gallon, but your point stands.

New comment by ZeroCool2u in "Launch HN: Didit (YC W26) – Stripe for Identity Verification"

ZeroCool2u — Wed, 11 Mar 2026 03:31:38 +0000

If you don't "have the full context on the Persona/Discord story" you should work on getting it.

It's literally the first thing that came to mind when I saw your post and not having a convincing/satisfying answer in direct relation to that catastrophe doesn't bode well for getting people to trust your brand. The rest of your answer is essentially the absolute minimum I'd expect from a business like this, but not sufficiently convincing.

Ask HN: What are some examples of your favorite product documentation?

ZeroCool2u — Tue, 10 Mar 2026 12:38:19 +0000

Comments URL: https://news.ycombinator.com/item?id=47322435

Points: 3

# Comments: 2

New comment by ZeroCool2u in "Yann LeCun raises $1B to build AI that understands the physical world"

ZeroCool2u — Tue, 10 Mar 2026 11:18:44 +0000

Regardless of your opinion of Yann or his views on auto regressive models being "sufficient" for what most would describe as AGI or ASI, this is probably a good thing for Europe. We need more well capitalized labs that aren't US or China centric and while I do like Mistral, they just haven't been keeping up on the frontier of model performance and seem like they've sort of pivoted into being integration specialists and consultants for EU corporations. That's fine and they've got to make money, but fully ceding the research front is not a good way to keep the EU competitive.

New comment by ZeroCool2u in "GPT-5.4"

ZeroCool2u — Thu, 05 Mar 2026 18:46:59 +0000

Yup, that was it. Didn't realize they're different models. I suppose naming has never been OpenAI's strong suit.

New comment by ZeroCool2u in "GPT-5.4"

ZeroCool2u — Thu, 05 Mar 2026 18:46:28 +0000

Ah yes, okay that makes more sense!

New comment by ZeroCool2u in "GPT-5.4"

ZeroCool2u — Thu, 05 Mar 2026 18:38:50 +0000

Frontier Math, GPQA Diamond, and Browsecomp are the benchmarks I noticed this on.

New comment by ZeroCool2u in "GPT-5.4"

ZeroCool2u — Thu, 05 Mar 2026 18:27:40 +0000

Bit concerning that we see in some cases significantly worse results when enabling thinking. Especially for Math, but also in the browser agent benchmark.

Not sure if this is more concerning for the test time compute paradigm or the underlying model itself.

Maybe I'm misunderstanding something though? I'm assuming 5.4 and 5.4 Thinking are the same underlying model and that's not just marketing.

New comment by ZeroCool2u in "President Trump bans Anthropic from use in government systems"

ZeroCool2u — Fri, 27 Feb 2026 21:12:59 +0000

Gift link: https://www.bloomberg.com/news/articles/2026-02-27/trump-ord...

Trump Orders US Agencies to Drop Anthropic After Pentagon Feud

ZeroCool2u — Fri, 27 Feb 2026 21:12:22 +0000

Article URL: https://www.bloomberg.com/news/articles/2026-02-27/trump-orders-us-government-to-drop-anthropic-after-pentagon-feud

Comments URL: https://news.ycombinator.com/item?id=47185682

Points: 20

# Comments: 3

New comment by ZeroCool2u in "LM Link: Use local models on remote devices, powered by Tailscale"

ZeroCool2u — Thu, 26 Feb 2026 16:16:11 +0000

I've been trying to use this all morning, but I keep getting 500/auth errors, even on a completely new device. Can't even login to my LM Studio account right now.

LM Link: Use local models on remote devices, powered by Tailscale

ZeroCool2u — Thu, 26 Feb 2026 16:13:46 +0000

Article URL: https://tailscale.com/blog/lm-link-remote-llm-access

Comments URL: https://news.ycombinator.com/item?id=47168021

Points: 2

# Comments: 0