251–260 of 352 posts
Earlier quoted context omitted.
https://techcrunch.com/2026/04/30/elon-musk-testifies-that-x... Make sure to stay updated!
Nothing against what I said in there. Distillation is far from enough to get to the frontier. Its at best a ramp (the most efficient one used by everyone ) that shortcut and saves millions of rl runs before a model moves.
Earlier quoted context omitted.
It still occupies a place on the Pareto frontier.
IMO Pareto frontier is a lie. At any given moment there's only one "best" strategy and my guess would be in most cases it is just using the best model.
It's a 20% discount on input and a 33% discount on output through at least November 21, 2026; the revised pricing schedule is now Model Input Cached input Cache writes Output gpt-5.6-sol $4.00 $0.40 $5.00 $20.00 gpt-5.6-terra $2.00 $0.20 $2.50 $12.00 gpt-5.6-luna $0.20 $0.02 $0.25 $1.20 So Sol is still 20x Luna, but much more appealing when compared to offerings from Anthropic and others.
It feels like a slightly more palatable version of what Anthropic has been doing, with their constant "use your free tokens before they expire next week!" campaigns. But it's feeling more and more ominous now, like they've hit the top of the demand curve and need to pull back prices to continue growing.
I am the type of coder that vibe codes - I talk to the agent about a problem, have it write a plan, red team the plan, and then implement the plan. These projects are things which haven’t really been done before, or if they have it’s not public or not in many places.
What I find is that Sol is hyper-left brained. Super focused on small details. When given a longer task with multiple steps it might go really hard on one of the early steps and it will validate, test, make safe, so much to the detriment of progressing the task within reasonable parameters for the project.
It also starts to sound crazy when you ask it for an update. It starts naming things in weird ways and the sentences don’t really make sense. It’s as if you’ve approached an engineer who has been hammering on something and he speaks to you in the lingo of his latest function, even though when you ask him for a status you are obviously asking about the whole project.
Fable on the other hand seems to remain coherent over time. It’s as if it remains aware of the longer run task. It’s got a bit more balance between left and right brain.
So OpenAI really need to find a balance between long term goal thinking and the very small task at hand.
For coders who apply Sol on specific functions or narrow tasks I’m a certain it is great. For me, a vibe coder, I need one that will be a bit more aware of the whole thing through these longer running tasks.
I think this is a move to get people off the subscription and move to API. The weekly usage is still awful altough it seems they're trying to fix it but I'm not hopeful.
Earlier quoted context omitted.
The naming schema actually tripped me up for a week or so. My brain's initial conception of the concepts was earth-relative, so I mapped it as: Sol = big, it's the sun Luna = medium, in-between sun and earth, space Terra = small, terrestrial
It tripped me up because I was going by distance. I thought Terra was the base model and Luna was the mid model because it’s further away.
But alas
Earlier quoted context omitted.
Fable 5 is just straight up a larger model - I'm guessing at this, but there is plenty of evidence online from people far more plugged in than I am. OpenAI is pursuing a strategy that yields greater operating margins and penetration of their model to developers. Fable's high cost makes it so premium that Anthropic has to reserve it for only the richest customers and corporate users. That's not a winning formula long…
Fable is available for $100 a month. If you're a working developer, you can pay that. I wouldn't really say it's "reserved for the richest customers".
The fact that AI models can be so easily distilled and replicated is such a stroke of luck. 10 or 15 years ago if one had asked me to envision a future where a private company invents artificial intelligence, I'd have thought for sure they'd have a massive moat, be very difficult to catch, and it would create an almost instant monopoly. Rather, it seems that selling intelligence might end up as a race to the bottom.…
Earlier quoted context omitted.
That’s not how it works though. Two training runs on the same data don’t produce the same weights. And if you want to modify the AI, you do so by fine tuning the weights not rerunning training. In every respect that matters, the weights are both the binary and the source code together.
If I want to remove censorship from your open-weights model, how do I do that?
Put in the work. This is akin to asking how to remove Rust from a Rust project; just because something is legally available to you doesn't mean you wont need to apply dome elbow grease, depending how deep the changes you want are, ablation, fine-tuning, or distillation are tools you can use to remove "censorship"
Earlier quoted context omitted.
They will have moat in the satellite launching business, which is not useful in the AI datacenter market. You can put AI chips in datacenters in the desert for far less than $100/kg. With lots of solar power available, the option to easily access your hardware and far less radiation issues. The datacenter in space story really only exists to make it possible for Musk to sell X to SpaceX and make more money from the I…
Yours is the only valid argument. If space datacenters are not competitive with terrestrial (in terms of megatokens/$ or whatever), then they will never work and SpaceX will fail. That's the whole ball game. But anyone who thinks they can predict those prices in ten years is wildly overconfident.
> then they will never work and SpaceX will fail
Space Datacenters eventually get quietly discontinued after finally killing the zombie brand at which point the only reaction is "huh, remember that?"
https://techcrunch.com/2026/08/21/teslas-solar-roof-is-dead-...