I’m working on some informations extraction from a scientifc text with the CHAT+ gpt4. with this sentence: “The first two dimensions concern evaluation novelty and valence” GPT 4 recognize a single dimension : “Assessment of novelty and valence” And GPT 3.5, the two dimensions: “-novelty: … -v…

You could try using a pinned model in the API to prevent any breaking changes. So for example, instead of calling gpt-4, you would call gpt-4-0613. You would write your code around a pinned model, which would ensure stability. Then before the pinned model gets deprecated, you hop to one of the la…

[image] curt.kennedy: So for example, instead of calling gpt-4, you would call gpt-4-0613. They are the same thing, merely aliases. Unfortunately there is currently no alternate AI model with function calling, and there are three months of alterations in performance, past what the date reflec…

[image] _j: Call gpt-4, see gpt-4-0613 in your response. I am aware of this, but I don’t assume the model quoted in the response is correct. You can only assume the model in your request is what you are running. Let me give you a hypothetical real world example: Suppose you build a giant A…

Of course I can show - but then I also can’t show, because the model has already been changed. Here’s about as clear as I can re-present Real Model Name Description gpt-4-0314 Initial release (some refinements past date possible) gpt-4-0613 (In June) As used in June? No longer exists gp…

@curt.kennedy Yep that’s what i planned to do. But i mean, can’t we expect to have at least the same level of interaction quality between each iteration of the same model? You see what i mean? With the api or with the chat+ service. If i take any software, the 1.10 version work “usualy” with the…

I would have trust issues too! These models are shifting over time, and over time they gain functionality, but can also lose functionality. The only way to guarantee consistency, for long durations, without worrying much about deprecation, is to run your own private model. I wish that OpenAI woul…

A “pinned model” version of gpt-4 beyond gpt-4-0314 doesn’t exist. That non-OpenAI language doesn’t accurately describe model invocation by API. Stable model name = recommended model => -0613 <= continuing updates I’ll let OpenAI’s text be the last word as we’ve both illustrated our notions and ar…

[image] _j: A “pinned model” version of gpt-4 beyond gpt-4-0314 doesn’t exist. Well, that sucks. They should release multiple “stable” pinned models periodically IMO. It only makes sense for developers who rely on system stability.

Chat+ GPT4 less smart than chat+ GPT3.5

ChatGPT

curt.kennedy September 23, 2023, 7:56pm 8

What you are showing is that the response is aliasing the floating update model with the pinned model.

So the only way to get the pinned model is to call it out in the request.

2 Likes

Topic		Replies	Views
Deprecation of chat-gpt-4o-latest Deprecations	8	2380	November 26, 2025
OpenAI should change roadmap Community chatgpt	7	243	March 16, 2026
Has the reasoning ability of the GPT 3.5 API dropped recently? API chatgpt , api	9	1172	December 25, 2023
I got the GPT-4 API, but I the model version is still the snapshot version of GPT-4 API	12	2900	December 17, 2023
Has regular gpt-4 model changed for the worse by any chance? Community gpt-4 , hallucinations	12	2058	April 23, 2025

Chat+ GPT4 less smart than chat+ GPT3.5

Related topics