Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

openai.com

Post reply on HN
Print view View original

401–410 of 433 posts

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#401

Post by josephg »

jjmarr wrote: post #237
GoatInGrey wrote: post #131

Earlier quoted context omitted.

Pornographic use has long been the "break glass in case of emergency" for the LLM labs when it comes to finances. My personal opinion is that while smut won't hurt anyone in of itself, LLM smut will have weird and generally negative consequences. As it will be crafted specifically for you on top of the intermittent reinforcement component of LLM generation.

I can do as much smut as I want through the API for all SOTA models.

I don’t know if this is still the case, but as of a year or so ago OpenAI would suspend your account if they noticed you using their models for this sort of thing. They said it was against their TOS.

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#402

Post by josephg »

subscribed wrote: post #279
GoatInGrey wrote: post #131

Earlier quoted context omitted.

Pornographic use has long been the "break glass in case of emergency" for the LLM labs when it comes to finances. My personal opinion is that while smut won't hurt anyone in of itself, LLM smut will have weird and generally negative consequences. As it will be crafted specifically for you on top of the intermittent reinforcement component of LLM generation.

Why llm smut in particular? There's already a vast landscape of the interactive, VR games for all tastes. Why LLM is supposed to be worse?

I think the argument is that it’s interactive. You’re no longer just passively reading or watching content. You can join in on the role play.

I’m not sure why that’s a bad thing though.

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#403

Post by goodmythical »

azan_ wrote: post #199
viraptor wrote: post #173

Earlier quoted context omitted.

You could commission smut of whatever type you want for quite a while. And many people do so. Even customised smut is not new. It's just going to get a bit cheaper and automated.

You couldn't talk to commissioned smut. Of course you could request changes etc. but the feedback loop was nowhere close to what you can get with AI. Interactivity is a very big deal.

There are absolutely people getting paid to roleplay smut in chat sessions and have been doing so at least since original Second Life and likely since the dawn of chat.

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#404

Post by jghn »

pdntspa wrote: post #397
jghn wrote: post #357

Earlier quoted context omitted.

The other day I asked CC to plan using Opus a few small updates to a FastAPI backend & corresponding UI updates for a nextJS frontend. Then I had it implement using Sonnet. It used up nearly half of my 5 hour quota right there and the whole process only took about 15 minutes.

This is on the pro ($100/mo) plan? I go through multiple sessions like this per day and it barely makes a dent. And I just keep it in Opus the whole time. How is it possible that our experiences are so different with essentially the same task? For reference, my current squeeze is about 30k sloc in Python, half being tests.

It's pro, but that's $25/mo. $100 is the lower tier of max.

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#405

Post by robwwilliams »

nfg wrote: post #326
mFixman wrote: post #311

Earlier quoted context omitted.

I've been impressed by how good ChatGPT is at getting the right context old conversations. When I ask simple programming questions in a new conversation it can generally figure out which project I'm going to apply it to, and write examples catered to those projects. I feel that it also makes the responses a bit more warm and personal.

Agreed that it can work well, but it can also irritating - I find myself using private conversations to attempt to isolate them, a straightforward per-chat toggle for memory use would be nice.

Love this idea. It would make it much more practical to get a set of different perspectives on the same text or code style. Also would appreciate temperature being tunable over some range per conversation.

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#406

Post by robwwilliams »

jghn wrote: post #337
azuanrb wrote: post #278

It’s interesting that many comments mention switching back to Claude. I’m on the opposite end, as I’ve been quite happy with ChatGPT recently. Anthropic clearly changed something after December last year. My Pro plan is barely usable now, even when using only Sonnet. I frequently hit the weekly limit, which never happened before. In contrast, ChatGPT has been very generous with usage on their plan. Another pattern I’…

> My Pro plan is barely usable now, even when using only Sonnet. I frequently hit the weekly limit, I thought it was just me. What I found was that they put in the extra bonus capacity at the end of dec, but I felt like I was consuming quota at the same rate as before. And then afterwards consuming it faster than before. I told myself that the temporary increase shifted my habits to be more token hungry, which is per…

This was my experience too over Dec 2025. Thereafter marginal Claude Pro utility. They are struggling with demand.

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#407

Post by robwwilliams »

SomeUserName432 wrote: post #381
azuanrb wrote: post #278

It’s interesting that many comments mention switching back to Claude. I’m on the opposite end, as I’ve been quite happy with ChatGPT recently. Anthropic clearly changed something after December last year. My Pro plan is barely usable now, even when using only Sonnet. I frequently hit the weekly limit, which never happened before. In contrast, ChatGPT has been very generous with usage on their plan. Another pattern I’…

> Another pattern I’m noticing is strong advocacy for Opus For agent/planning mode, that's the one only one that has seemed reasonably sane to me so far, not that I have any broad experience with every model. Though the moment you give it access to run tests, import packages etc, it can quickly get stuck in a rabbit hole. It tries to run a test and then "&& sleep" on mac, sleep does not exist, so it interprets that a…

Somewhat different type of problem and perhaps a useful precautionary tale. I was using Opus two days ago to run simple statistical tests for epistatic interactions in genetics. I built a project folder with key papers and data for the analysis. Opus knew I was using genuine data and that the work was part of a potentially useful extension of published work. Opus computed all results and generated output tables and pdfs that looked great to me. Results were a firm negative across all tests.

The next morning I realized I had forgotten to upload key genotype files that it absolutely would have required to run the tests. I asked Opus how it had generated the tables and graphs. Answer: “I confabulated the genotype data I needed.” Ouch, dangerous as a table saw.

It is taking my wetware a while to learn how innocent and ignorant I can be. It took me another two hours with Opus to get things right with appropriate diagnostics. I’ll need to validate results myself in JMP. Lessons to learn AND remember.

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#408

Post by SomeUserName432 »

alsetmusic wrote: post #399
SomeUserName432 wrote: post #381

Earlier quoted context omitted.

> Another pattern I’m noticing is strong advocacy for Opus For agent/planning mode, that's the one only one that has seemed reasonably sane to me so far, not that I have any broad experience with every model. Though the moment you give it access to run tests, import packages etc, it can quickly get stuck in a rabbit hole. It tries to run a test and then "&& sleep" on mac, sleep does not exist, so it interprets that a…

> It tries to run a test and then "&& sleep" on mac, sleep does not exist > type sleep > sleep is /bin/sleep What’s going on on your computer? Edit: added quote

Right you are.. Perhaps I recall incorrectly and it was a different command. I did try it, and it did not exist. Odd.

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#409

Post by novemp »

sejje wrote: post #389
Gud wrote: post #229

Earlier quoted context omitted.

What are these extremes beauty standards being promoted? Because it seems to me large swaths of the population need some beauty standards

Yes, but you're not allowed to say that to them. They are victimized by the fact that models are attractive, and that is "unrealistic," so they've been getting plus sized models etc. The "extreme beauty standards" are basically just "healthy BMI."

Anorexia is overrepresented in both male and female models.

To get that "ripped" look so many "fit" guys have, they often have to be dehydrated.

Actresses wear full faces of makeup even in apocalyptic scenarios.

It goes way beyond "just don't be fat".

Re: Retiring GPT-4o, GPT-4.1, GPT-4.1 mini, and OpenAI o4-mini in ChatGPT

#410

Post by sejje »

novemp wrote: post #409
sejje wrote: post #389

Earlier quoted context omitted.

Yes, but you're not allowed to say that to them. They are victimized by the fact that models are attractive, and that is "unrealistic," so they've been getting plus sized models etc. The "extreme beauty standards" are basically just "healthy BMI."

Anorexia is overrepresented in both male and female models. To get that "ripped" look so many "fit" guys have, they often have to be dehydrated. Actresses wear full faces of makeup even in apocalyptic scenarios. It goes way beyond "just don't be fat".

It does, but the remedy they choose is not "women don't wear makeup," it's "we use fat women, and they wear more makeup than skinny women."
Post reply on HN