Technology

I don't care how rough the estimate is, LLMs are using insane amounts of power, and the message I'm getting here is that the newest incarnation uses even more.

BTW a lot of it seems to be just inefficient coding as Deepseek has shown.

I have an extreme dislike for OpenAI, Altman, and people like him, but the reasoning behind this article is just stuff some guy has pulled from his backside. There's no facts here, it's just "I believe XYX" with nothing to back it up.

We don't need to make up nonsense about the LLM bubble. There's plenty of valid enough criticisms as is.

By circulating a dumb figure like this, all you're doing is granting OpenAI the power to come out and say "actually, it only uses X amount of power. We're so great!", where X is a figure that on its own would seem bad, but compared to this inflated figure sounds great. Don't hand these shitty companies a marketing win.

I think AI power usage has an upside. No amount of hype can pay the light bill.

AI is either going to be the most valuable tech in history, or it's going to be a giant pile of ash that used to be VC capital.

Bit of a clickbait. We can't really say it without more info.

But it's important to point out that the lab's test methodology is far from ideal.

The team measured GPT-5’s power consumption by combining two key factors: how long the model took to respond to a given request, and the estimated average power draw of the hardware running it.

What we do know is that the price went down. So this could be a strong indication the model is, in fact, more energy efficient. At least a stronger indicator than response time.

Fucking Doc Brown could power a goddamn time machine with this many jiggawatts, fuck I hate being stuck in this timeline.

There's such a huge gap between what I read about GPT-5 online, versus the overwhelmingly disappointing results I get from it for both coding and general questions.

I'm beginning to think we're in the end stages of Dead Internet, where basically nothing you see online has any connection to reality.

And an LLM that you could run local on a flash drive will do most of what it can do.

For reference, this is roughly equivalent to playing a PS5 game for 4 minutes (based on their estimate) to 10 minutes (their upper bound)

How the hell are they going to sustain the expense to power that? Setting aside the environmental catastrophe that this kind of "AI" entails, they're just not very profitable.

I don’t buy the research paper at all. Of course we have no idea what OpenAI does because they aren’t open at all, but Deepseek's publish papers suggest it’s much more complex than 1 model per node… I think they recommended like a 576 GPU cluster, with a scheme to split experts.

That, and going by the really small active parameter count of gpt-oss, I bet the model is sparse as heck.

There’s no way the effective batch size is 8, it has to be waaay higher than that.

Of course there are comments doubting the accuracy, which by itself is valid, but they are merely doing it to defend AI. IMHO, even at a fifth of the estimates, we’re talking humongous amounts of power, all for a so-so search engine, half arsed chatbots and dubious nsfw images mostly. And let’s not forget: it may be inaccurate and estimates are TOO LOW. Now wouldn’t that be fun?

Isn't this the back plot of the game, Rain World? With the slug cats and the depressed robots stuck on a decaying world when the sapient, organic species all left?

That's alright. When they've got a generation of people who can't even hold a conversation without it, let alone do a job, that price increase will drop that energy use pretty rapidly.

This bubble needs to pop, the sooner the better.

that's a lot. remember to add "-noai" to your google searches.

40Wh or 18Wh which is it?

That's my old gaming PC running a game for 2min42sec-6minutes ... Roughly.

The last 6 to 12 months of open models has pretty clearly shown you can substantially better results with the same model size or the same results with smaller model size. Eg Llama 3. 1 405B being basically equal to Llama 3.3 70B or R1-0528 being substantially better than R1. The little information available about GPT 5 suggests it uses mixture of experts and dynamic routing to different models, both of which can reduce computation cost dramatically. Additionally, simplifying the model catalogue from 9ish(?) to 3, when combined with their enormous traffic, will mean higher utilization of batch runs. Fuller batches run more efficiently on a per query basis.

Basically they can't know for sure.