GPT-5.2 vs GPT-5.4
As of Oct 14, 2026, GPT-5.2 leads GPT-5.4 in 0 of 8 shared categories.
GPT-5.2
#14 Best current rank: #14 in Math, as of Oct 14, 2026.
GPT-5.4
It used to be. Was SOTA in General LLM.
| Category | GPT-5.2 | GPT-5.4 |
|---|---|---|
| General LLM | #74 · #74 in General LLM Margin 50.34013605442177 LMArena leaderboard dataset · Oct 13, 2026 | #28 · Was SOTA · General LLM Margin 19.04761904761905 LMArena leaderboard dataset · Oct 13, 2026 |
| Reasoning | #70 · #70 in Reasoning Margin 48.63013698630137 LMArena leaderboard dataset · Oct 13, 2026 | #29 · #29 in Reasoning Margin 20.54794520547945 LMArena leaderboard dataset · Oct 13, 2026 |
| Math | #14 · #14 in Math Margin 28.888888888888886 Epoch AI benchmarking hub · Dec 11, 2025 | #9 · #9 in Math Margin 17.777777777777757 Epoch AI benchmarking hub · Mar 5, 2026 |
| Science | #26 · #26 in Science Margin — Epoch AI benchmarking hub · Dec 11, 2025 | #16 · #16 in Science Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
| Coding | #— · Not enough data Margin — Epoch AI benchmarking hub · Dec 11, 2025 | #— · Not enough data Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
| Coding › Web dev | #67 · #67 in Coding › Web dev Margin 61.6822429906542 LMArena leaderboard dataset · Oct 13, 2026 | #55 · #55 in Coding › Web dev Margin 50.467289719626166 LMArena leaderboard dataset · Oct 13, 2026 |
| Long context | #80 · #80 in Long context Margin 54.42176870748299 LMArena leaderboard dataset · Oct 13, 2026 | #34 · #34 in Long context Margin 23.12925170068027 LMArena leaderboard dataset · Oct 13, 2026 |
| Vision | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 9, 2026 | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 9, 2026 |
| Factuality | #65 · #65 in Factuality Margin 49.25373134328358 LMArena leaderboard dataset · Oct 13, 2026 | #6 · Was SOTA · Factuality Margin 5.22388059701494 LMArena leaderboard dataset · Oct 13, 2026 |
| Abstraction (ARC) | #24 · #24 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Dec 11, 2025 | #10 · #10 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
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.
General LLM: As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in General LLM at #28 against #74.
Reasoning: As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Reasoning at #29 against #70.
Math: As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Math at #9 against #14.
Science: As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Science at #16 against #26.
Coding › Web dev: As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Coding › Web dev at #55 against #67.
Long context: As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Long context at #34 against #80.
Factuality: As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Factuality at #6 against #65.
Abstraction (ARC): As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Abstraction (ARC) at #10 against #24.
GPT-5.2 or GPT-5.4?
As of Oct 14, 2026, GPT-5.2 leads GPT-5.4 in 0 of 8 shared categories.
GPT-5.2 or GPT-5.4 for General LLM?
As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in General LLM at #28 against #74.
GPT-5.2 or GPT-5.4 for Reasoning?
As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Reasoning at #29 against #70.
GPT-5.2 or GPT-5.4 for Math?
As of Oct 14, 2026, GPT-5.4 leads GPT-5.2 in Math at #9 against #14.