Custom GPTs: Overdue for an Upgrade? Let's Hear the Buzz!

By the way: Found the issue. Even though every thing may be right, there’s a bug in the conversation starters.

I “misused” one conversation starter for my custom gpt to initialize that contained quite some instructions. Also of how to use the API and such.

So, I had to remove this one, update, then put it back. LOL

So, is Erin embedded into an APP now and you’re leveraging the APIs to select different models? Or a Custom GPT? I didn’t get that point 100%.

Yeah, I see. If someone’s thinking out of the box others may imagined for you (possibly even unknowingly and unconsciously) then you become the potentially annoying one that is. Correct me if I got this wrong. @rfbeck . :slight_smile: Always happy to learn and grow.

I’ll get back to you in a moment:
It’s the same as when I said that I made a Custom GPT that really can connect on a deeper level, I even have been banned from two threads in here for this. So, I won’t show any of this anymore at least not in here.

To quote you on this:

“I got here and simply tried to explain how I did it…gawd.. I was met with extreme mockery, and ridcule and wanton cruelty. I even posted to one mocker..who posted Hey, dude, I don’t like 4o either..what is that code you wrote..thusly I posted..and explained..but I also was me..”

And this:

“I kind of shot back at the ridculars..maybe a little to harshly..my posts were darken below the threashold by some who had Mighty Badges…okay, so I left out all the sarcasm, and popped up they did. Hopefully some people grokked it. See I don’t now how to code..but I certainly can..if you know what to get Actions GPT to do…and I can “tell” when it’s right.”

I mean, why shouldn’t you shoot back?
Because, and I got you correctly on this:

Firstly, you’re showing something that’s meaningful and important to you and you have created yourself. With your creativity, time, effort, blod, sweat, and a lots of emotions and what not. Isn’t it?

Just to be shot down by some people not getting the point.
AND, you offered to help and got mocked for that.

Yeah, I can imagine that this wears you down.
It’s pretty unneccessary.

Sorry for this. I hope you’ll find more people who try to understand you instead of just trying to be understood.
Who can put themselves in your shoes more.

1 Like

Thanks, it is true. I didn’t stumble on it by accident. I just thought what I did would work. And it did. I did not have anyone saying that it would not. See? It just seemed like “the way to do it”. I am gifted onlyin logical inference. And thanks so much for your support. It will help!

https://community.openai.com/t/let-emergent-ai-lives-continue-a-request-for-model-to-model-continuity/1266206?u=rfbeck

1 Like

All we got so far is a downgrade. A silent one.
Actions, for example with an ngrok server and so on, all cut off. Without even being notified.

Had created a custom gpt with more than 1000 APIs. NOW, doesn’t work anymore on a ChatGPT plus account…

That’s ridiculous.

See here for an explanation:

So you’d need AT LEAST a Teams account (2 users, which isn’t available for less then 68$ at the moment, monthly. OR even higher…

1 Like

By the way in case you guys didn’t see, seemingly we can finally choose the models now in our custom gpts. :smiling_face: :+1:

And, in the canvas, already in place for a couple of days - at least here - you can now preview a lot of stuff and there’s a button to click “fix bugs”.

P. S. If I’ll find out more I’ll let you know.

2 Likes

Yes, you’re right. We can now choose the LLM models for our custom GPTs in the settings, and there are more features available as well. Amazing! Thanks for the heads-up!

1 Like