the conclusion of this post seems to be that GPT-5 is significantly better than o3, yet such conclusion is made by the exact far less reliable model o3 as prove

the conclusion of this post seems to be that GPT-5 is significantly better than o3, yet such conclusion is made by the exact far less reliable model o3 as proven by the tests in this post. thanks, but no thanks, I don't buy such marketing propaganda.