I don't much care about benchmarks. Talk obvious, commonsensical increases in utility to me.

There wasn't that much utility in old, unreliable AI.

I remember their capabilities like so:

GPT-3 could produce convincing looking texts, sometimes.

GPT-4 was somewhat smarter and would give more accurate answers. At that time image understanding was released, wasn't it?

Then with GPT-5 we have a reasoning model, another jump in capabilities.

Now, compare GPT-6 Astra with its ability to implement software, work on long horizon tasks, and visual understanding.

If you haven't seen the (massive) increases in utility I'd wager you haven't been using the tools much.