Gemini 3.1 Pro Preview vs GPT-6 Astra
As of Oct 15, 2026, Gemini 3.1 Pro Preview leads GPT-6 Astra in 4 of 9 shared categories: General LLM, Reasoning, Long context, and Factuality.
Gemini 3.1 Pro Preview
Yes. Yes on Math (best per dollar). Not the overall leader.
GPT-6 Astra
Yes. SOTA in Math and co-SOTA in Vision.
| Category | Gemini 3.1 Pro Preview | GPT-6 Astra |
|---|---|---|
| General LLM | #15 · Was SOTA · General LLM Margin 7.9787234042553195 LMArena leaderboard dataset · Oct 13, 2026 | #26 · #26 in General LLM Margin 13.829787234042556 LMArena leaderboard dataset · Oct 13, 2026 |
| Reasoning | #14 · Was SOTA · Reasoning Margin 7.9787234042553195 LMArena leaderboard dataset · Oct 13, 2026 | #24 · #24 in Reasoning Margin 13.297872340425528 LMArena leaderboard dataset · Oct 13, 2026 |
| Math | #3 · Near-SOTA · Math Margin 4.444444444444443 Epoch AI benchmarking hub · Feb 19, 2026 | #1 · SOTA · Math Margin 0 Epoch AI benchmarking hub · Sep 3, 2026 |
| Science | #7 · #7 in Science Margin — Epoch AI benchmarking hub · Feb 19, 2026 | #2 · #2 in Science Margin — Epoch AI benchmarking hub · Sep 3, 2026 |
| Coding | #— · Not enough data Margin — LMArena leaderboard dataset · Oct 13, 2026 | #2 · Near-SOTA · Coding Margin 13.63636363636364 Terminal-Bench Harbor Hub · Oct 6, 2026 |
| Coding › Agentic | — | #2 · Near-SOTA · Coding › Agentic Margin 13.63636363636364 Terminal-Bench Harbor Hub · Oct 6, 2026 |
| Coding › Web dev | #63 · #63 in Coding › Web dev Margin 49.6 LMArena leaderboard dataset · Oct 13, 2026 | #2 · Near-SOTA · Coding › Web dev Margin 0.7999999999999972 LMArena leaderboard dataset · Oct 13, 2026 |
| Long context | #12 · Was SOTA · Long context Margin 6.382978723404236 LMArena leaderboard dataset · Oct 13, 2026 | #25 · #25 in Long context Margin 13.297872340425528 LMArena leaderboard dataset · Oct 13, 2026 |
| Vision | #25 · Was SOTA · Vision Margin 68.42105263157896 Epoch AI benchmarking hub · Feb 19, 2026 | #1 · Co-SOTA · Vision Margin 0 Epoch AI benchmarking hub · Sep 3, 2026 |
| Factuality | #21 · #21 in Factuality Margin 12.79069767441861 LMArena leaderboard dataset · Oct 13, 2026 | #29 · #29 in Factuality Margin 17.441860465116278 LMArena leaderboard dataset · Oct 13, 2026 |
| Abstraction (ARC) | #2 · #2 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Feb 19, 2026 | #1 · #1 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Sep 3, 2026 |
Gemini 3.1 Pro Preview: $4.5 per 1M blended tokens.
GPT-6 Astra: $20 per 1M blended tokens.
Gemini 3.1 Pro Preview: closed weights.
GPT-6 Astra: closed weights.
General LLM: As of Oct 15, 2026, Gemini 3.1 Pro Preview leads GPT-6 Astra in General LLM at #15 against #26.
Reasoning: As of Oct 15, 2026, Gemini 3.1 Pro Preview leads GPT-6 Astra in Reasoning at #14 against #24.
Math: As of Oct 15, 2026, GPT-6 Astra leads Gemini 3.1 Pro Preview in Math at #1 against #3.
Science: As of Oct 15, 2026, GPT-6 Astra leads Gemini 3.1 Pro Preview in Science at #2 against #7.
Coding › Web dev: As of Oct 15, 2026, GPT-6 Astra leads Gemini 3.1 Pro Preview in Coding › Web dev at #2 against #63.
Long context: As of Oct 15, 2026, Gemini 3.1 Pro Preview leads GPT-6 Astra in Long context at #12 against #25.
Vision: As of Oct 15, 2026, GPT-6 Astra leads Gemini 3.1 Pro Preview in Vision at #1 against #25.
Factuality: As of Oct 15, 2026, Gemini 3.1 Pro Preview leads GPT-6 Astra in Factuality at #21 against #29.
Abstraction (ARC): As of Oct 15, 2026, GPT-6 Astra leads Gemini 3.1 Pro Preview in Abstraction (ARC) at #1 against #2.
Gemini 3.1 Pro Preview · GPT-6 Astra
Gemini 3.1 Pro Preview or GPT-6 Astra?
As of Oct 15, 2026, Gemini 3.1 Pro Preview leads GPT-6 Astra in 4 of 9 shared categories: General LLM, Reasoning, Long context, and Factuality.
Gemini 3.1 Pro Preview or GPT-6 Astra for General LLM?
As of Oct 15, 2026, Gemini 3.1 Pro Preview leads GPT-6 Astra in General LLM at #15 against #26.
Gemini 3.1 Pro Preview or GPT-6 Astra for Reasoning?
As of Oct 15, 2026, Gemini 3.1 Pro Preview leads GPT-6 Astra in Reasoning at #14 against #24.
Gemini 3.1 Pro Preview or GPT-6 Astra for Math?
As of Oct 15, 2026, GPT-6 Astra leads Gemini 3.1 Pro Preview in Math at #1 against #3.