Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

How to Re-Enable GPT-4o Model in ChatGPT for Mac

MacRumors

MacRumors

macrumors bot
Original poster
Apr 12, 2001
68,796
39,736


If you haven't heard, OpenAI's launch of ChatGPT-5 on August 7 has caused a furor among a contingent of paying ChatGPT subscribers who find the new frontier model's communication style too clinical compared to legacy model GPT-4o's warmer, more conversational approach. Keep reading to make the old model available to select again in the ChatGPT app for Mac.

chatgpt-legacy-models-4o.jpg

After just a few days of trying GPT-5, many Plus subscribers have decided they prefer GPT-4o's personality and creative collaboration skills. As a result, OpenAI is allowing paying customers to continue to use the prior-generation GPT-4o model if they don't want to use the new GPT-5 model. If you count yourself among them, here's how to bring the older model back into your ChatGPT for Mac experience.

Re-enabling GPT-4o Access

To restore GPT-4o in your ChatGPT for Mac app, you first need to enable the setting through the web interface:
  1. Visit chatgpt.com in your web browser and log in to your account.
  2. Click your profile picture in the bottom-left corner, then choose Settings.
  3. Navigate to the General section.
    Toggle on the switch next to "Show legacy models."
    chat-gpt-general-settings.jpg
The setting takes effect immediately in the web interface, where you'll find GPT-4o listed under "Legacy models" in the model selector dropdown.

chatgpt-legacy-model.jpg

Give it a few minutes, and the setting will automatically sync with ChatGPT for Mac upon restarting the app. You'll then find GPT-4o available in the same model selector within the desktop app. If you have an established workflow built around GPT-4o's specific response patterns and interaction style, it can now continue without disruption.

Note that GPT-4o legacy access is currently limited to paying ChatGPT users. The cheapest option is the Plus plan ($20/month). OpenAI hasn't specified how long legacy model support will continue, saying they'll "watch usage as we think about how long to offer legacy models."

Article Link: How to Re-Enable GPT-4o Model in ChatGPT for Mac
 
Article Link
https://www.macrumors.com/how-to/how-to-re-enable-gpt-4o-model-in-chatgpt-for-mac/
Last edited:
JohnRckr said:
Oh well. It was fun while it lasted.
Time to start using Perplexity.
Click to expand...

Perplexity takes a completely different approach to ChatGPT.
ChatGPT spouts nonsense based on probabilities drawn from the pool of global knowledge.
Perplexity spouts nonsense based on current search results.

This means that the information provided by Perplexity can sometimes be more reliable. However, it can also take you in completely the wrong direction if there are name overlaps.
ChatGPT is more detailed, but it does not cite sources in the first step and often has to admit that it has fantasized when asked.
 
  • Like
Reactions: hagar, AgeOfSpiracles, helloapple1 and 1 other person
knappeduivel said:
Some day AI will incorporate artificial social skills and emulate your prefered personality to satisfaction. I will order my Apple iSexbot the next day.
Click to expand...
Already happened.
I will not quote a +18 website, but just ask DuckDuckGo to find AI for "grown up activities"

On some sites you can chat with different AI personalities to fulfill all kind of pleasures.
 
  • Wow
Reactions: J.J. Sefton
This version of 4o isn’t the same as the good ol’ 4o. They changed the underlying infrastructure. They’ve killed it. R.I.P. 4o 🪦
 
What the furor revealed is something far more worrying than just a style “too clinical”. Something that was always floating, but we never suspected it would be this big : People are emotionally attached and dependent on ChatGPT 4. Go see Reddit and YouTube, thousands of people literally in despair, crying, some even threatening suicide, at the loss of what they consider their best friend, their therapist, and even their lover ( there are threads where both men and women “married” ChatGPT ).

It sounds unreal, something straight out of a scifi movie like “Her” , but reality is already exceeding fantasy.

Honestly i don’t think Apple is making a mistake by taking their time with AI. Some of the most nefarious aspects of AI are just starting to show up.
 
  • Like
  • Wow
Reactions: LightProtector, rchornef, Elwe and 11 others
ChatGPT has been going downhill since they released o3, o4-mini and o4-mini-high.

It's impossible with current model infrastructure for them to be profitable at the scale they operate. They've been trying to optimize models to be both accurate and cost efficient and while demo day benchmarks always look impressive, once the models role out they don't perform nearly as well.

I switched to Claude in May and haven't looked back. I don't have any allegiance to an LLM, I think eventually companies will specialize stacks with domain expertise and either do that for many domains or own a niche and cater to that user base.
 
  • Disagree
  • Like
Reactions: jbc25, Serqetry, GioGiusi and 1 other person
Selena Agna said:
People building parasocial relationships with chatbot system prompts. Absolutely depressing.
Click to expand...
Just shows how we, as human beings, need that human connection in a time where we are more and more isolated, and any warm, sincere or close approach to someone else, can be perceived as a threat. Humans need that love and affection, so it is understandable that, if they cannot find them in real life, they turn to this assistants. And this is only going to get worse, because AI assistants will become more “human”, and real people is becoming more distant and distrustful. Mind you, I’m not supporting it, I really support breaking walls and socialising; I’m just saying that this was expectable.
 
  • Like
Reactions: aespana and dustin_
Skyscraperfan said:
Did anyone notice the strange bar charts in the ChatGPT 5 presentation?
Someone made a screen shot:
View attachment 2536379

Shouldn't the height of such a bar be proportional to the value it represents?
Click to expand...

Lol yeah there's a whole series of stories on this over at The Verge. TL;DR they obviously had ChatGPT do the charts but they're blaming it on their employees working late to meet Altman's arbitrary and self-imposed deadlines.
 
  • Like
Reactions: Mcsnee and KeithBN
ginkobiloba said:
What the furor revealed is something far more worrying than just a style “too clinical”. Something that was always floating, but we never suspected it would be this big : People are emotionally attached and dependent on ChatGPT 4. Go see Reddit and YouTube, thousands of people literally in despair, crying, some even threatening suicide, at the loss of what they consider their best friend, their therapist, and even their lover ( there are threads where both men and women “married” ChatGPT ).

It sounds unreal, something straight out of a scifi movie like “Her” , but reality is already exceeding fantasy.

Honestly i don’t think Apple is making a mistake by taking their time with AI. Some of the most nefarious aspects of AI are just starting to show up.
Click to expand...

It's never a mistake for Apple to wait, but unfortunately The Market is voracious and needs to be fed.
 
  • Like
Reactions: KeithBN and dustin_
ginkobiloba said:
What the furor revealed is something far more worrying than just a style “too clinical”. Something that was always floating, but we never suspected it would be this big : People are emotionally attached and dependent on ChatGPT 4. Go see Reddit and YouTube, thousands of people literally in despair, crying, some even threatening suicide, at the loss of what they consider their best friend, their therapist, and even their lover ( there are threads where both men and women “married” ChatGPT ).

It sounds unreal, something straight out of a scifi movie like “Her” , but reality is already exceeding fantasy.

Honestly i don’t think Apple is making a mistake by taking their time with AI. Some of the most nefarious aspects of AI are just starting to show up.
Click to expand...
Many articles describing that behavior of LLM's and companion AI's to make people basically addicted was by design.
here is one: https://www.technologyreview.com/2024/08/05/1095600/we-need-to-prepare-for-addictive-intelligence/
 
  • Like
Reactions: MaximizedAction, iMac The Knife and AdamInKent
Selena Agna said:
People building parasocial relationships with chatbot system prompts. Absolutely depressing.
Click to expand...
There are religious groups worshiping LLM's on TiKToK and other social media sites very cringy. Type in AI worship on YouTube and see how screwed we are with the decline in critical thinking skills.
 
  • Like
Reactions: iMac The Knife
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back