181–190 of 352 posts
Earlier quoted context omitted.
Altman specifically has said in an interview that I listened to once that he envisions AI being as cheap as electricity.
I hope it's a good bit cheaper than that, I pay close to $400/mo for electricity and I'm in no way interested in paying anything like that for AI.
I live in Germany where people won't stop whining about electricity prices, and I pay 75€/mo.
Earlier quoted context omitted.
Altman specifically has said in an interview that I listened to once that he envisions AI being as cheap as electricity.
I hope it's a good bit cheaper than that, I pay close to $400/mo for electricity and I'm in no way interested in paying anything like that for AI.
My prediction is that this becomes permanent. There is no good reason to be much more expensive than opus. At $4/$20 they are roughly at parity. Making 2/10 permanent would be a killer move and make a strong argument against open-weight. For the sake of the open weight ecosystem I hope they do not.
Earlier quoted context omitted.
Nemotron Super is sort of open source in the sense that Nvidia provides almost everything you need to replicate it from scratch. Of course it’s performance is not exactly stellar but it could be a good starting point for other research teams.
Nemotron is okay. Better than Olmo.
Earlier quoted context omitted.
OpenAI could still have a significant moat. ChatGPT occupies most consumers’ minds when they think about AI and has become a household name. Google won because search became a habit-forming product people grew accustomed to using. Bing was once effectively indistinguishable from Google Search, yet still failed to achieve mass adoption because users had already become accustomed to “Googling” things. The same could be…
Difference is that it was free to google/bing search. Ai prompting costs money. If I run out of tokens on ChatGPT of course I will try Claude. I never ran out of Google searches so no reason to try Bing
At some point Google gets suspicious of your persistent searches and makes you solve captchas and puts cooldowns on your searches.
The top comment on this thread was about AI models being easily distilled being a stroke of luck. This should not be surprising at all. Every new students spends tiny fractions of time learning knowledge that took many lifetimes to discover. This fundamental to the progress of intelligence and understanding. It should not be surprising that AI can be distilled. It's the logical method of training; I would hope that e…
This argument is exactly why we should not anthropomorphise models.
You are comparing the way a human brain learn with training a statistical model. You can't just "this is like learning so don't be surprised".
It takes a child one minute to learn how to open a padlock. Teach that to a robot with your analogies.
My prediction is that this becomes permanent. There is no good reason to be much more expensive than opus. At $4/$20 they are roughly at parity. Making 2/10 permanent would be a killer move and make a strong argument against open-weight. For the sake of the open weight ecosystem I hope they do not.
Earlier quoted context omitted.
Distillation a pretty well documented technique that actually pre-dates LLMs https://arxiv.org/pdf/1503.02531 Here is a project that guides you through it if you want to prove to yourself that it works https://github.com/arcee-ai/DistillKit
That distillation exists isn't the question. It's about evidence this is an active force in competition in LLMs.