frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

GitLab 19.6.0 is out and still vulnerable to Ruby2Shell and NGINX2Root

10•hoechst•1h ago•3 comments

Anthropic (ANT) stock lists at nearly $2T valuation

4•_hmj•19m ago•0 comments

Ask HN: When do we stop switching and start resonating?

2•DiogenesOfYerev•31m ago•0 comments

Ask HN: When was the last time you thought "that's bad API design"?

4•Mitchem•2h ago•0 comments

Rant: Gemini 4 does not make any sense

4•sambejk•2h ago•5 comments

Does Liquidity Drive the Stock Market? We Tested 22 Years of Fed Data

3•evgeniymezenov•2h ago•0 comments

Ask HN: How do you manage the curse of rabbit hole?

7•BOOSTERHIDROGEN•6h ago•6 comments

Ask HN: If AI tools are so good, why isn't your bug backlog empty?

103•ex-aws-dude•2d ago•115 comments

Why is China treated as threat-actor but not India?

6•jcarmackdaddy•2h ago•13 comments

Ask HN: Cheapest CPU Vms

13•mrkn1•16h ago•8 comments

Ask HN: When does your app need its own UI vs. using it via MCP?

7•amoorthy•16h ago•6 comments

Interesting Learnings Working for a Startup

19•Jessiejia11•1d ago•5 comments

Developing Browser Extensions

3•LayerTake•20h ago•3 comments

Ask HN: Tips for transitioning from web development to embedded/firmware?

4•llmblockchain•15h ago•4 comments

Tell HN: /tmp is hogging your RAM on Ubuntu 26

8•boldi•15h ago•2 comments

Claude Gets Hella Paranoid

19•noduerme•1d ago•11 comments

I open-sourced an AI loop for RISC-V CPU design months prior to OpenAI

9•fsbonetto•17h ago•1 comments

Ask HN: What are you working on? (November 2026)

305•david927•1w ago•1113 comments

Codeberg seems to have vanished from DNS

58•samip537•1d ago•21 comments

Ask HN: How do you reach the flow with vibe coding?

5•varlock•1d ago•11 comments

Ask HN: How do you use AI to get more relevant content?

4•anoojb•1d ago•5 comments

You've reached the end!

Open in hackernews

Rant: Gemini 4 does not make any sense

4•sambejk•2h ago
On November 18th, Google released another model, this one called Neon. Making it the fourth model in just 2 months.

OpenAI makes sense: Luna-Terra-Sol-Astra. Moon is smaller than earth, earth is smaller than sun and I can forgive them for not knowing that sun is also a star.

Anthropic is even better: Haiku-Sonnet-Opus-Mythos/Fable. Haiku is an extremely short poem, Sonnet is a bit longer, Opus is very long and Mythos is extremely long. Ok, Fable does not exactly really fit in there, but they came up with it after they only had the three models, so we had time to adjust. Also, Claude told me: Fable is basically Mythos with extra safety, and a fable is a story with a moral.

Now, I want you to think back to those high school chemistry classes. Which is bigger neon or carbon or hydrogen or argon? Impossible right? The correct answer, hydrogen, carbon, neon, argon, doesn't even match with model capabilities. To make things worse, there is no consistency, they are combining noble gases with other non-metals. You can't even say they went for easy names, helium, is not in there as an example. They really should have gone with just noble gasses: helium, neon, argon and xenon (skipping krypton to not confuse with kryptonite)

I'm glad that Google is sort of back in the race, but they should really invest in someone who is not AI-pilled to come up with these names.

PS don't get me started on Meta, wtf is a glimmer?

Comments

endean-eng•2h ago
When they dropped Argon with an emphasis on low hallucination and named it after a Noble gas, I was really hoping that was going to be a theme.

Then Carbon made sense since it was the most versatile and could build (bond to) almost anything. Fine, works I guess.

And then Hydrogen dropped. Hydrogen is light and versatile, so that makes sense...kind of.

Along came the Helix leak. Never heard of that element before.

And now we have Neon. Which has a hallucination rate between Hydrogen and Carbon...but is named after a noble gas.

Yep. Agreed. Not sure what's going on anymore.

sambejk•1h ago
they apparently are going to release Krypton which is at Fable level. I hope they just retire the hydrogen and carbon models and then have a reasonable line-up... but that would be too much to expect from them.
muzani•2h ago
I get the impression these models were made by different, competing teams in Google.
kbrannigan•2h ago
You are onto something. Probably has the most training data and brain to create a superior product, too big to unite to actually deliver.

My bet is microsoft,

sambejk•1h ago
At 5000 people, Deepmind is by far the biggest AI lab. Would make sense if they just let them compete. But I think it is giving them too much credit.