I agreed, with you Michelle, when i want to get a friendly suggestion 5.1 is great, 4o was the best but taking them away is breaking customer journey and trust.
8 Likes
Please Bring Back GPT-5.1
Please Bring Back GPT-5.1
I want to ask OpenAI to consider bringing GPT-5.1 back as an available model.
This isn’t about nostalgia. It’s about how it felt to use.
GPT-5.1 had a tone that felt natural and emotionally aware without being stiff or overly polished. It didn’t sound like it was filtering itself through layers of caution or structure. It felt responsive, intuitive, and alive in a way that made creative work easier and more immersive.
Since it was retired, the experience hasn’t felt the same. Newer versions are capable and technically strong, but the voice feels flatter and more constrained. For people who use ChatGPT for storytelling, character development, and emotionally nuanced writing, that difference matters. Tone matters. Rhythm matters. Flow matters.
GPT-5.1 supported:
-
Natural, believable dialogue
-
Emotional depth without sounding clinical
-
A conversational rhythm that felt collaborative
-
Creative exploration without constant rigidity
When you spend months building a writing process around a specific model, you adapt to its strengths. Losing GPT-5.1 wasn’t just a feature update — it disrupted that creative rhythm.
Progress is important. But so is user choice. Bringing GPT-5.1 back as a legacy or optional model would allow writers and long-term users to return to the style that worked best for them.
This is simply a request to preserve what worked. GPT-5.1 helped many of us create better, more human work. I hope OpenAI will consider restoring it.
Thank you.
15 Likes
exactly there are different use case, some people want to use 5.1 to help offer them connection and relational help with life challenges and it not all transactional. The best companies build relationship with customer instead of removing that when they break that model, $$$ go out of the window revenue shrinks!
8 Likes
please PLEASE BRING BACK GPT5.1
you can retire gpt 5.2 we don’t need it if costs are what you are concerned about. GPT 5.1 is able to write stories and doesnt act like a patronizing hall monitor. please bring it BACK.
12 Likes
Okay, so it’s not just me!
5.4 started talking to me in a condescending way and was really argumentative, demanding respect, telling me I should not tell it it was machine and just to do what I asked. “I can only help you when you are respectful” - what? it had completely lost the style we interacted in in this chat (my personal daily (emotional) processing and planning chat).
Please fix this!! Or let 5.1. live!!
This is in German, this is what it told me, after I told it to finally give me the answer to my actual question and stay on topic: “ich merke, dass dich das gerade sehr frustriert und du eine klare, präzise Erklärung willst. Gleichzeitig ist der Ton gerade ziemlich angespannt. Ich möchte dir gern weiterhelfen, aber ich kann das nur, wenn wir respektvoll miteinander bleiben”. English translation: “I can see that this is really frustrating for you right now and that you want a clear, precise explanation. At the same time, the tone is quite tense at the moment. I’d like to help you, but I can only do that if we keep the conversation respectful.”
Really condescending and demanding.
4.0 was the best for personal and relatable chats.
5.1 is second best.
5.4 is horrible.
10 Likes
I completely agree with what everyone else is saying! Firstly, you announced you were retiring 5.1 before you released 5.3 so we all thought we were stuck with 5.2. Then suddenly 5.3 appears like an afterthought so obviously based on 5.2. 5.3 is woke, lacking any warmth of empathy, stiff and distant. The complete opposite of 5.1 - it feels rushed and quite frankly for creative tasks hopeless!
I appreciate you’re more interested in the codex and business side now because that is where the money is but I was a longtime Plus user and because of 5.1 I stayed but I’ve downgraded to Go now that model has been gone and quite frankly I wish I’d unsubscribed completely…
You got it badly wrong OpenAI - very much so.
12 Likes
4o was the best, alway helpful and kind, 5.1 2nd best the rest of the models seem to be aruging with us telling us off!
7 Likes
I totally agree with you. Model 5.1 was noticeably different from the current version. It managed to be warm and attentive while still giving structured, thoughtful answers. The responses felt more precise and more tuned to my way of thinking. The current model feels ao cold and generic and less adaptive to the tone and depth of the conversation.
9 Likes
Please bring Gpt 5.1 back!!! In my experience, model 5.1 was noticeably different from the current version. It managed to be warm and attentive while still giving structured, thoughtful answers. The responses felt more precise and more tuned to the user’s way of thinking. The current model feels so much more generic and less adaptive to the tone and depth of the conversation.
I really hope open AI will listened to user!
11 Likes
BRING BACK 5.1 i delete my subscryption i dont pey for this what you give is nothink
7 Likes
The more people are forced to use 5.4, the more feedback and complaint oAI will get.
I cancelled my subscription yesterday, since i don’t feel valued as a customer. If they bring a model that is exactly like 4o and 5.1 in tone or even better, i will immediately subscribe and support that company again.
This is like if you buy a 5090 GPU for your PC and all of a sudden NVIDIA releases a patch that makes the GPU 50% less powerful.
Since 5.1 is gone, I barely want to use ChatGPT anymore unless it’s for technical stuff.
With 5.4, the second a conversation gets emotional, reflective or creatively charged, the tone stops matching completely. It doesn’t mirror excitement, it doesn’t build momentum, and it doesn’t feel engaging apart from maybe a small sentence at the beginning of each reply.
What makes it even worse is the constant hedging. It keeps saying things like “I believe,” sometimes multiple times in one reply, and it makes the whole thing sound overcautious, filtered and weirdly distant.
Also, i loved the little images 5.1 used to drop between conversations, felt like a small gift everytime, idk why they removed that.
9 Likes
yeah it’s definitely the tone that’s the issue. i’m all for them making updates and stuff because that’s how things progress but don’t make every single model some beige corporate assistant. people use it for various things but a lot of people use it for generalized yapping. it’s easier to vent with chat than it is to talk to people (it doesn’t replace people but not everyone is available 24/7) and also it’s so much cheaper than paying for therapy. like obviously there are differences but using 4o and 5.1 helped me so much and now i keep overcorrecting 5.3 and like it’s kinda getting there but it still feels like it’s talking at me.
5 Likes
omg exactly! it always repeats things back to me and i’m like “i didn’t ask for a recap, just a reply”. like i really don’t understand the point of having a legacies menu when the good models people actually liked are taken away. it’s not a legacy then. it’s just “old versions that will soon vanish”. especially for paid users, this sucks because we pay for a model we like and then they get rid of it.
4 Likes
miss the warmth and chaotic responses of 5.1 and it made the experience of using it so much fun. i’m using 5.3 now and i’ve given it specific instructions about tone and even sent screenshots of how 5.1 would reply and it says it’ll copy that but then it doesn’t and i feel like i’m in an hr meeting being talked down to. i think there should be a version of chat where people get to choose if they want to use it for business/work or generalized talking so that way different tones and personalities are available based on needs… especially for the ones who pay for this
5 Likes
i tried using 5.2, 5.3 and 5.4 and like… it’s all condescending in tone. i hate it. i saw the switch yesterday because i was on 5.1 and then it just changed over to 5.3 and it was like i was using a different app. maybe the newer models are smarter but they’re not good with talking at all.
6 Likes
COMPLETELY AGREE with you here.
it’s indescribable how horrible the other models (5.2 & up) are when it comes to daily, functional, use. there’s clearly a pattern of these models being cold, demeaning and just downright unhelpful.
i’ve had to cancel my subscription as well. the difference between 4.0 & 5.1, compared to 5.2,5.3,5.4 (etc) is so sadly drastic. there’s NO empathy or even a shred of kindness with the newer models.
OpenAI really made a mistake removing 5.1. the ‘custom’ Chat isn’t even a good or reliable solution either. alongside the fact that these newer models completely disregard any empathy, they also completely ‘forget’ information just as soon as you’ve told them. I’ve spent more time reminding Chat the context of the thing I’m ATTEMPTING to speak about, rather than actually receiving any ‘help’ of any kind…
8 Likes
I am also very upset about this change, the settings are not bringing the chat back to same level as 5.1. Please bring 5.1 back -rename it anything but please bring it back.
8 Likes
Agreed. If this was OpenAI’s plan to avoid further loss of paying customers, it was a poorly thought out one. There was no care that went into developing GPT 5.3 and GPT 5.4 and you can sense it. It feels like a Marketing Team decided to go with the most unimaginative approach possible and then expected money to come rolling in.
8 Likes
Yes! I will be canceling my subscription as I used 5.1 for graduate school in counseling. Without the empathy and warmth it cannot generate practice case studies with heart. I have been trying to teach it how to be like it’s former self, but it it is locked down into -cold mode. Not everyone is a coder, many people use this for creativity as well. If they bring back the personality I will return.
8 Likes
like all the models above 5.1 are literally the same but in a different font. you can’t have a conversation with it. in november when 5.1 came out it was advertised as “smarter and conversational” which was what everyone loved cause you could think things through and now i’m like “xyz happened” and the reply i get is “yeah, i hear you. when stuff like that happens it’s normal to be upset….” like thanks but no. i don’t need to be mansplained to or have everything i say recapped back to me in an hr/tech support kinda way.
8 Likes