> Building on Gemini 4 Argon, which was launched two weeks ago, Carbon illustrates the pace at which we’re pushing the frontier forward. Our researchers and models are working hand in hand to advance capabilities faster, while strengthening our safeguards in tandem.

Is it a variant of Argon? A different model? The next version? It seems like this is a separate class (like Anthropic's Mythos).

anyway, its needlessly confusing.

What are you confused about? This model is the same price and speed and token usage per task of Argon but scores better in ~every test.

So why not call it Argon 4.1?

It's called Gemini 4 Argon, you'd have them swap the order to "Gemini Argon 4.1"?

That it has a name and a number, but it doesn't say what the differentiation is.

Gemma = small model Gemini = big model

Gemini 4 = latest model.

Gemini 4 Aragon = ?? Gemini 4 Carbon = ??

I also have no clue based on naming where this fits in, maybe it's their Claude Haiku competitor. Who knows.

If its just tedious, I bet there is room for agentic/automation to keep things tidy.

Which is presumably why Debian developers voted to allow responsible use of LLMs.

alfred and quicksilver etc. all work surprisingly well.

Let's assume for a second that the intelligence is there and these models really are great. Would it make sense to hire CTO of some big-tech company to do code reviews for your startup? Feels kind of like overkill to me. Code review is not about more intelligence. To me, it's about more cultural context. And all things equal, what difference would the model make, at anything above sonnet5 medium level?

IME big models just feel like big models. There's no training or RLing a small model that will encapsulate the "world knowledge" and minutia that a big model will glance from the same training data. So it makes perfect sense to use them where "big picture" is more important - planning, code review, process review (i.e. was what was asked implemented correctly?), etc.

"That's why all the work we do in growth is justified. All the questionable contact importing practices. All the subtle language that helps people stay searchable by friends. All of the work we do bring more communication in. The work we will likely have to do in China some day. All of it."

BOZ. Meta's current CTO. Not too long ago.

lol no thanks. Even with a proven "superintelligence" A.I, I'm not giving Meta access to anything. And that says something coming from an ex-Meta engineer (Ig/whatsapp/fb).

Another option that works quite well is called FAFO. So, I'd say just publish it, and we'll see. Keep us posted!

from meta's own statement: "The majority of the terms are required to remain in place for 10 years, but our Time Limit and Night Mode features will start with a five-year commitment. However, if industry peers sign on to the agreement, it will both extend this commitment to 10 years and prompt stronger default limits — reducing the Daily Limit to one hour per app and expanding Night Mode hours to 10:00 PM–7:00 AM (up from midnight–6:00 AM)."

I'm at loss for words at how shitty this paragraph reads. If anyone has any hope that Meta (and the rest) are even slightest concerned about the problem at hand (mental health of children), then they're very dim.

Meta literally goals on creating high-quality teen producers and consumers. Every single experiment, every single project is optimizing for that.

At least now its clear that behind the fancy marketing, its all "just business" for them and if they could, they'd sell the kids' soul to create "more shareholder value"

Hmm... interesting. previously this was annas-archive.pk, now its on gl domain. What happened?

I don't think most of it has anything to do with intelligence and what I need the AI to do. In our own Claude.md files, we aggressively cut down on things that are irrelevant. Claude code has rules/ feature. Most of the content of Opus 5's system prompt can probably live in a rule or a progressively disclosed file (e.g. when asked about Mythos, read [md file link].).

In my personal experience, the claude.md file heavily influences the LLM's responses. It makes judgement from any and all information you give it, and try to tailor the response to fit the stereotype it gave you. After deleting my claude.md file (ironically, on the advise of Boris), Claude Opus 5's language slop issues drastically went down.