Ugh, people are still saying the Codex limits are more generous. They're not, Claude's are over 2x higher, have been for months! [1] It's just that Claude uses far more tokens, 2-3x is common. Except sometimes GPT will use just as many or even go into a compact loop and then your quota is gone, little headroom for hard tasks.
That website seems to suggest that Opus 5 spends ~57 cents per task, while GPT 5.6 Sol spends ~49 cents per task? That ratio doesn't feel quite right to me. Artificial Analysis says Opus 5 High costs nearly ~3x as much as GPT 5.6 Sol High for a given task: https://artificialanalysis.ai/models/comparisons/claude-opus...
Yeah I subscribe to both and watch the numbers too, and it drives me nuts
Who wants to actually watch anyways, rather than worry about it my team just created our own harness that prioritize usage + intelligence and assigns work out (and records token usage..)
We do see them here occasionally on US import cars. (NL) Tend to be older models like those long and wide models with benches in front and back and very loose suspension.
Here in the UK, older cars are often exempt from various safety standards. Otherwise, recent imported vehicles have to comply with e.g. headlights dipping the wrong way which is common as we drive on the left.
> I'm surprised. But it shouldn't be up to Tesla; it's illegal and they never get on the road in countries with proper regulations.
The USA doesn't mandate red turn signals. They can be red or amber. Plenty of BMWs with amber turn signals. Although the current one has red turn signals.
I'd say on modern cars it's less bad, the turn signal is usually a strip, and the brake light is larger, joined with a 3rd brake light.
But on one of those old Ford vans, with incandescent red brake/turn signals with no/often broken 3rd light: you have to wait a second before you can determine if the person is braking / indicating.
NY state has 30% tint law, but my Uber driver last night was more than that.
I’ve heard tint being enforced with other violations, but i believe window tint analyzers are in short supply or costly, and they have to share. Which leads to longer roadside detainments.
I'm surprised NY allows tinting of the whole front windshield. Here in AZ we're bound by the (stupid) AS-1 line. Tinting here would help more than in NY.
It’s cultural; style over function—-unless the function is more traffic deaths (for which I have no numbers of traffic deaths involving tinted windows, I freely admit. I have had uncomfortable interactions with tinted cars on the streets and highways, so they have earned my permanent enmity).
The trouble is both the cops and bully boys like tinted windows the same as acculturated Captain America and the Red Skull.
I urge you to reconsider your beliefs. You are missing something important because you are thinking in terms of low-dimensional statistics. Deep learning doesn't just fit data, it finds features (abstractions) of the data.
I asked the original commenter to confirm my understanding of what they were saying.
I find it a fascinating alternative view to what is largely well understood (your counter-point).
The thing that stood out is the comment that it’s an approximation of the original data generator (humanity). Early approximations were poor (GPT 2-3, to an extent GPT-4).
I’m not so sure I can reject the hypothesis that such an approximation can be found.
IME using 5.6 Luna and DS V4 Flash, I notice that although they are excellent at programming, even Opus-like in the way they try to debug, the thing they are worst at is inferring user intent and making good decisions with little information. They are absolutely terrible at that, will misinterpret small wording ambiguities. I suspect that's an ability you can't add with RL training, that it requires the depth of understanding from vast pre-training.
I think this is the best and most useful way to measure model intelligence. In my experience it's what really sets apart the capable models from the best. A small model can be RL trained to be extremely good at programming or narrow problem solving for its size (eg 5.6 Luna, DS4 Flash, Qwen 3.6 27B), but even Luna is IME comparatively awful at understanding intent and making good decisions with limited guidance.
If doesn't correspond cleanly. I can see why you draw the link, because LZ compression will replace words with symbols but BPE is a non-contextual entropy encoding while LZ is contextual and adaptive and that makes it very different. I think BPE actually has more in common with Huffman encoding.
That conception of knowledge is interesting, but I think using the label 'knowledge' for it is very problematic, it's too far from common definitions. The fact that you have to carve out an exception for mathematics already shows there's a problem. Because if maths, shouldn't thought experiments also produce new knowledge? You're excluding special and general relativity. It seems to me that what the concept actually describes is "information about the world".
> The fact that you have to carve out an exception for mathematics already shows there's a problem.
I don't see how that's an exceptional case. Mathematical theories may not correspond to physical objects, but they still have to be tested for consistency, logical validity etc. As for relativity, one could argue that Einstein's theories were indeed not "knowledge" until their predictions were confirmed (like, say, the gravitational lensing thing). The Michelson–Morley experiment preceded Einstein, but it had to be checked that the theory was consistent with that, and so on and so forth.
There is a distinction between a compressor for a fixed dataset and one for an unknown population from which we have a sample. The optimal compressor for the sample may be the single best guess for the population, but that's not what Solomonoff induction does. It begins with a prior that allows all possible programs, and it never assigns all probability to the single optimal compressor, so it has no problem with the all-zeroes example.
But the Hutter prize (of which I'm a big fan) is for ever-more-optimal compressors, and in fact many of the solutions don't generalise to other input data without stripping out various tricks.
I think it's better to think of the hutter prize as a challenge to see how close people can get to efficiently approximating AIXI without the exponential cost of AIXItl. The fact that winners don't generalize well is just that they're not great approximations, because the behavior is different in the limit.
The winner generalize really well. They're just incredibly slow.
Most of the neural compressors are trying to squeeze every last bit out by sacrificing runtime. You can speed them up a fair bit and they're pretty good. fast-cmix with a small priming corpus will probably be about twice as compact as something like lzma or brotli.
> For me it’s the massive amount of resources it takes to produce and run one
It's amazing that LLM pretraining is both extremely data inefficient at learning concepts and cognitive functions from the training data compared to humans, while actually being quite efficient at learning facts, memorising things seen just a few times.
I used to likewise think that the resources required to run large transformers were absurd, but the architectures are far more efficient now than 3 years ago and I underestimated just massive the parallelisation advantage of transformers is, how many TFLOPS effective you can get. You can already run amazingly decent LLMs on PCs and phones.
I generally agree with you, but my view has shifted from "we need to augment or replace LLMs" to it there being far more efficient algorithms possible but it not actually being necessary for fulfilling most goals.
You misread. "Pain and suffering" not "death". Of all the pain and suffering in the world, a vast amount of it really is our own fault. Famines and wars shouldn't happen. And if you see a country border with poverty and one side and prosperity on the other you can't say that was the only possibility.
[1] https://devforth.io/agents-for-code/?sortby=monthly-value And I can confirm the numbers, I subscribe to both and watch the numbers
reply