Mirror mirror on the wall, who’s the best model of all?

Let me add my two cents on the international nightmare of the model picker!

Since the launch of GPT-5, Open AI has drastically simplified the model picker, to the point that initially there wasn’t really one anymore. If you ask me, that was the right move. One of the things I never liked about the UI of ChatGPT was that model picker, mostly because the names of the models never made any sense.

o3 was better than 4, but not necessarily same as 4.5? What?

I was super happy when it became just 5. Nice and simple and let’s be honest: if we are aiming for AGI, shouldn’t these models be clever enough to figure out independently the best tool to pick to present the answer to my prompt? Seems like a pretty basic task to me!

It seems I was in the minorty as pretty much the same day OpenAI made a big U turn and brought back yet another type of model selector! Even less descriptive and more cryptic. But that wasn‘t enough as they also brought back legacy models for the 4o afficionados….

So in the span of 24 hrs they want from super minimalistic and simple to more complex than before GPT5.

Sam Altman tweeted (Xed?) this about the update:

You can now choose between “Auto”, “Fast”, and “Thinking” for GPT-5. Most users will want Auto, but the additional control will be useful for some people. […] 4o is back in the model picker for all paid users by default. If we ever do deprecate it, we will give plenty of notice. Paid users also now have a “Show additional models” toggle in ChatGPT web settings which will add models like o3, 4.1, and GPT-5 Thinking mini. 4.5 is only available to Pro users—it costs a lot of GPUs.

Seems like commitment issues from OpenAI took over right after this less than stellar launch.

Over-hyped and under-delivered. We probably need GPT6 to interact with GPT5 picker to solve this once and for all…