Codex 5.4 vs 5.5 pricing and quality (opens in new tab)

You can get very close results to GPT 5.5 by using GPT 5.4 with a highly detailed prompt. I ran a small test to check this properly. I generated the same technical content into summaries using both GPT 5.4 and GPT 5.5, across four different prompt detail levels (Low to XHigh). Then I asked ChatGPT to rank all 8 outputs blindly, without giving it any scoring categories or guidelines — so my own preferences wouldn’t influence the result. Here’s how it turned out: Rankings (1 = best): GPT 5.5 XH...