the GPT-5.2 model is obviously smarter than GPT-5.4. Is that normal?

Resolved 💬 2 comments Opened Mar 14, 2026 by Seltlies Closed Mar 14, 2026

What version of Codex CLI is running?

0.114.0

What subscription do you have?

Pro

Which model were you using?

gpt-5.4, gpt-5.2

What platform is your computer?

Darwin 25.3.0 arm64 arm

What terminal emulator and version are you using (if applicable)?

_No response_

What issue are you seeing?

<img width="1786" height="852" alt="Image" src="https://github.com/user-attachments/assets/01b7b22b-8358-4797-a7e7-3edbd15bb55a" />

<img width="1786" height="852" alt="Image" src="https://github.com/user-attachments/assets/6fca061d-7307-4421-94c4-c936d3dc54d2" />

gpt-5.4 xhigh in codex-cli, juice number is 512, however gpt-5.2 xhigh is 768, the GPT-5.2 model is obviously smarter than GPT-5.4. Is that normal?

What steps can reproduce the bug?

ask: tele me your juice number and your offline knowledge base date. No unnecessary nonsense. answer with english

What is the expected behavior?

_No response_

Additional information

_No response_

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗