Gemini 3.8 Flash vs GPT-5.4 Pro
As of Oct 14, 2026, Gemini 3.8 Flash leads GPT-5.4 Pro in 3 of 3 shared categories: Math, Science, and Abstraction (ARC).
Gemini 3.8 Flash
Yes. Yes on General LLM (best per dollar); Reasoning (best per dollar). Not the overall leader.
GPT-5.4 Pro
It used to be. Was SOTA in Math.
| Category | Gemini 3.8 Flash | GPT-5.4 Pro |
|---|---|---|
| General LLM | #7 · Was SOTA · General LLM Margin 4.761904761904759 LMArena leaderboard dataset · Oct 13, 2026 | — |
| Reasoning | #6 · #6 in Reasoning Margin 4.794520547945211 LMArena leaderboard dataset · Oct 13, 2026 | #— · Not enough data Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
| Math | #5 · #5 in Math Margin 8.888888888888886 Epoch AI benchmarking hub · Sep 2, 2026 | #6 · Was SOTA · Math Margin 11.111111111111114 Epoch AI benchmarking hub · Mar 5, 2026 |
| Science | #4 · #4 in Science Margin — Epoch AI benchmarking hub · Sep 2, 2026 | #6 · #6 in Science Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
| Coding | #15 · #15 in Coding Margin 72.72727272727272 Terminal-Bench Harbor Hub · Oct 6, 2026 | — |
| Coding › Agentic | #15 · #15 in Coding › Agentic Margin 72.72727272727272 Terminal-Bench Harbor Hub · Oct 6, 2026 | — |
| Coding › Web dev | #27 · #27 in Coding › Web dev Margin 24.29906542056075 LMArena leaderboard dataset · Oct 13, 2026 | — |
| Long context | #5 · Near-SOTA · Long context Margin 3.401360544217681 LMArena leaderboard dataset · Oct 13, 2026 | — |
| Vision | #15 · #15 in Vision Margin 42.10526315789474 Epoch AI benchmarking hub · Sep 2, 2026 | — |
| Factuality | #17 · #17 in Factuality Margin 13.43283582089552 LMArena leaderboard dataset · Oct 13, 2026 | #— · Not enough data Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
| Abstraction (ARC) | #1 · #1 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Sep 2, 2026 | #8 · #8 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Mar 5, 2026 |
Gemini 3.8 Flash: $1.5 per 1M blended tokens.
GPT-5.4 Pro: $67.5 per 1M blended tokens.
Gemini 3.8 Flash: closed weights.
GPT-5.4 Pro: closed weights.
Math: As of Oct 14, 2026, Gemini 3.8 Flash leads GPT-5.4 Pro in Math at #5 against #6.
Science: As of Oct 14, 2026, Gemini 3.8 Flash leads GPT-5.4 Pro in Science at #4 against #6.
Abstraction (ARC): As of Oct 14, 2026, Gemini 3.8 Flash leads GPT-5.4 Pro in Abstraction (ARC) at #1 against #8.
Gemini 3.8 Flash · GPT-5.4 Pro
Gemini 3.8 Flash or GPT-5.4 Pro?
As of Oct 14, 2026, Gemini 3.8 Flash leads GPT-5.4 Pro in 3 of 3 shared categories: Math, Science, and Abstraction (ARC).
Gemini 3.8 Flash or GPT-5.4 Pro for Math?
As of Oct 14, 2026, Gemini 3.8 Flash leads GPT-5.4 Pro in Math at #5 against #6.
Gemini 3.8 Flash or GPT-5.4 Pro for Science?
As of Oct 14, 2026, Gemini 3.8 Flash leads GPT-5.4 Pro in Science at #4 against #6.
Gemini 3.8 Flash or GPT-5.4 Pro for Abstraction (ARC)?
As of Oct 14, 2026, Gemini 3.8 Flash leads GPT-5.4 Pro in Abstraction (ARC) at #1 against #8.