cesarvarela

Proof of work vs. intelligence.

That's still the point though.

I get that you're saying the model might get a higher score because of luck, but at the same time you want the model to be "lucky".

I wish Argon were as lucky as Opus.

The scores look similar to how it feels to use Argon: great at world knowledge, not so good at coding compared to Opus.

It's hard to understand the phased rollout of a model that isn't particularly sota, unlike Mythos a few months ago.

Argon has been so disappointing that I don't know what to think. It feels like an older-generation model, except for its world knowledge.

Medium or High? Depending on the task, you might want to switch between them.

Subscriptions provide thousands of dollars' worth of tokens if measured at API prices. While the value was once absurd, it is now slightly lower but still much cheaper.

You are paying API prices, and 6.1 Terra gets you 90% of what 6.1 Sol gets you at half the price.

It is: Fable 5.5 will be better than Opus 5.5. Opus 6 will be better than Fable 5.5, Fable 6 will be better than Opus 6, etc.

Just ask Opus and Kimi to build a game about something random, like "make a Mario Kart-inspired game." Opus gives better results most of the time. That doesn't mean you can't get your particular job done with either of these models.

Yeah, but it's not good enough. Ask Opus and Astra to make a landing page about lemon pies or whatever. Opus will get you a better result almost every time.

I wish Opus 5.5 had an alternative just for the sake of not having a monopoly, but I have yet to see a model live up to its benchmark scores except for Fable and Opus; there is something about Anthropic that no other lab has.

Just as some people can run two marathons back-to-back while others get tired after walking around the block, it has to be that way.

Very nice visual and sound experience, congrats!

Yeah, but also way more people are getting portraits than before the camera was invented—there’s a bit of a Jevons paradox there.

The issue is time. Even if DeepSeek matches Opus on 9 out of 10 tasks, failing on the tenth can cost you more in lost time than you saved in money.

It's funny that most of the AI industry is built around the assumption (which is most likely true) that it is not possible to run SOTA models on current consumer hardware.

Imagine if someone managed to run an Astra- or Fable-level model on a 5090 at reasonable speeds.

I think a book is like one completion of the mind behind it, so in a way, this is just distillation.

I'll try that since I'm on the 100 plan and they disabled the 200 one.

It looks like the optimal pattern is to have Astra as the orchestrator and Sol as the implementer. Same as with Fable and Opus.

Claude code still uses it internally.

I noticed this with DeepSeek Flash and DigitalOcean. They are doing something seriously wrong when serving that model; we need something like an SLA, but for intelligence; it's almost fraudulent behavior.

I trust Apple to do better here; for example, after about ten years, no one has caught up to Face ID yet.

A little bit of this, a little bit of that, we know the average, but we don't know how much time each of us has on this earth.

The transition effect from closed to open is so chef's kiss; the little things that set Apple apart.

So, on the one hand, we have AGI; on the other, the release page is returning 500s.

I'm curious about the Omniscience index because OpenAI has been lagging Anthropic on it.

I find it funny that while these releases are a technological miracle, the charts in the doc use tiny fonts and are hard to read. Goes with the idea that coding might be solved, but taste isn't.

I wish he kept a lower profile. His public behavior risks overshadowing the work SpaceX is doing, work that could benefit humanity in ways we can’t yet fully anticipate. I wish we had more Elons pushing us toward a "future that looks like the future," as he said.

Space exploration might help solve poverty, though. It will create new occupations, as new technology and discoveries always do.

Once you switch to another model (I've been playing with Grok, for example), you'll notice the mental overhead of reading Claude slop. It's like having an AC or a vacuum making noise in the background, then it stops and you feel relief.