It's not available to new users (I think there is a "karma" threshold but not sure about the exact number) and you need to to a direct link to the comment (e.g. click the time in the comment header) to see the option.
People also love type 2 fun. It's not fun in the moment, but you're happy that you did it.
If your work is type 1, more power to you. A lot more falls under type 2's umbrella. I find writing to be type 2 more often than not. Making complicated designs is often not fun in the moment. I like exercise, but sprint workouts are type 2.
A complete accounting of fun types has to include Type 0 fun as well: fun when it's happening, but not fun later. Drugs, gambling, crime, and most traditional vices fall in this category.
The problem is that the task you've defined "split up a task to create a chain of agents" has changed dramatically in just the last six months, nevermind the last two years.
You're wasting effort and teaching an obsolete technology if you try to make primary/secondary education too topical. Students can learn how to decompose a task and how to think critically without ever touching a Large Language Model.
Probably. But the difference is the marginal cost of selling an Adobe CS license, Avid Media Composer, or the other costly software I bought at a steep discount is pretty much nothing. When you discount inference you lose money.
Pulling the plug on K-12 on the other hand, seriously, can’t happen fast enough.
True! But Adobe gives cloud compute and some gen AI credits with their edu license. Autodesk does too! They both lose money on that proposition, but clearly not thousands or tens of thousands of dollars per user.
No reasonable person would be confused by use of baking soda as an ingredient in cooked food (reasonable) vs the addition of baking soda after cooking as an adulterant.
Did you really mean to say 4.5? Gpt 4.5 used to cost $75/$150 per million tokens input/output. And it did not even seem to be that good to justify that. I would not expect many people were using it, and I doubt that "expanding to india" was what killed it (if it was that useful/popular they would have kept the api, or keep it for higher end subscriptions).
If anything it should have been no1 in the "openAI graveyard" website.
India in this context is a synecdoche for scaling consumer vs Anthropic's more enterprise-y route, but yes that's pretty much why we didn't get 4.5 with reasoning. Without reasoning, 4.5 had no future.
From Sam Altman himself:
> We had this big GPU crunch. We could go make another giant model. We could go make that, and a lot of people would want to use it, and we would disappoint them. And so we said, let’s make a really smart, really useful model, but also let’s try to optimize for inference cost. And I think we did a great job with that.
4.5 scaled into a unified reasoning model would have been an incredible model. It beat GPT-5 on accuracy and hallucinations without reasoning (!)
It just wouldn't have worked for powering things like ChatGPT Go's rollout and loginless chatgpt.com, so they dropped it.
(And if you want, you could argue it's the compute crunch that didn't let them do both... but Anthropic had to make the same choices at the time and went in the other direction.)
This all sounds like pure speculation to me. GPT4.5 was ok but not spectacular. The whole marketing was based on "vibes" and how interacting with it "felt more natural" etc. If there was actual use case for this model, I do not see why it would not be just offered for higher end subscriptions or through API. Other expensive models at the time, eg o1/o3 pro, were not served in the free tier, but only in paid subscriptions and apis, but that one did have use cases, so they did keep it at the time, until they prob took a more unified approach with their models. So I do not see why they could not have done something similar with 4.5 if it was an actually good model.
And I am not sure that Altmat's statements are worth taking into account. His statements are more about marketing and turning things in his favour rather than speaking the truth.
+1 Deciding what to write is the critical step. You can get it with careful typing, but it's harder because you can type fast enough to skip that step.
Replace financial sector with big tech and overexposure to CDOs to overexposure to large promised data centers build outs and you have the right answer.
That has the benefit of letting you create/edit/export the model in a single application instance in a single workflow that is easy with practice.
reply