# GPT-5.2 vs GPT-5.4

As of Oct 15, 2026, GPT-5.2 leads GPT-5.4 in 0 of 8 shared categories.

GPT-5.2: $4.8125 per 1M blended tokens.

GPT-5.4: $5.625 per 1M blended tokens.

GPT-5.2: closed weights.

GPT-5.4: closed weights.

- Which to pick, General LLM: As of Oct 15, 2026, GPT-5.4 leads GPT-5.2 in General LLM at #30 against #92.
- Which to pick, Reasoning: As of Oct 15, 2026, GPT-5.4 leads GPT-5.2 in Reasoning at #32 against #88.
- Which to pick, Math: As of Oct 15, 2026, GPT-5.4 leads GPT-5.2 in Math at #9 against #14.
- Which to pick, Science: As of Oct 15, 2026, GPT-5.4 leads GPT-5.2 in Science at #16 against #26.
- Which to pick, Coding › Web dev: As of Oct 15, 2026, GPT-5.4 leads GPT-5.2 in Coding › Web dev at #60 against #73.
- Which to pick, Long context: As of Oct 15, 2026, GPT-5.4 leads GPT-5.2 in Long context at #40 against #100.
- Which to pick, Factuality: As of Oct 15, 2026, GPT-5.4 leads GPT-5.2 in Factuality at #6 against #79.
- Which to pick, Abstraction (ARC): As of Oct 15, 2026, GPT-5.4 leads GPT-5.2 in Abstraction (ARC) at #10 against #24.

As of Oct 15, 2026, GPT-5.2 is ranked #14 in Math (since Oct 15, 2026; sources Humanity's Last Exam; methodology v1.0).

As of Oct 15, 2026, GPT-5.4 is a former state-of-the-art holder in General LLM (since Apr 10, 2026; sources LMArena Text (style-controlled, overall); methodology v1.0).