Gemini 3.1 Pro benchmark results
14 results on 10 benchmarks from 4 sources.
Also listed as: gemini-3.1-pro-preview · gemini-3.1-pro-preview_high · Gemini 3.1 Pro Preview · gemini-3.1-pro-preview-customtools · Gemini 3.1 Pro Preview (thinking high) · Gemini-3.1-Pro · Gemini 3.1 Pro (Preview)
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1487 | 17 | Overall, style control on; 124675 votes; 95% CI 1484–1490“gemini-3.1-pro-preview” | Official leaderboard | Oct 8, 2026 |
| LMArena WebDev | 1447 | 71 | Overall; 24249 votes; 95% CI 1442–1452“gemini-3.1-pro-preview” | Official leaderboard | Oct 8, 2026 |
| LMArena Vision | 1279 | 27 | Overall, style control on; 45558 votes; 95% CI 1274–1285“gemini-3.1-pro-preview” | Official leaderboard | Oct 2, 2026 |
| GPQA Diamond | 94.4% | 8 | Epoch AI run, accuracy | Independent run | Aug 6, 2026 |
| MMLU-Pro | 91.2% | 1 | Overall accuracy; data source: TIGER-Lab“Gemini-3.1-Pro” | Official leaderboard | Mar 11, 2026 |
| SWE-bench Verified | 75.6% | 12 | Epoch AI run (Epoch scaffold) | Independent run | Feb 24, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 95.6% | 43 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 6, 2026 |
| Humanity's Last Exam | 46.4% | 3 | Compiled by Epoch AI | Third-party compilation via Epoch AI | Feb 19, 2026 |
| τ²-bench | 26.0% | 17 | banking_knowledge domain, pass^1; effort: high; submitted by Sierra; v1.0.1“Gemini 3.1 Pro Preview” | Official leaderboard | May 5, 2026 |
| Terminal-Bench 2.0 | 80.2% | 6 | Terminal-Bench 2.0; agent: TongAgents | Official leaderboard via Epoch AI | Feb 19, 2026 |
| Terminal-Bench 2.0 | 78.4% | 10 | Terminal-Bench 2.0; agent: Forge Code | Official leaderboard via Epoch AI | Mar 2, 2026 |
| Terminal-Bench 2.0 | 78.4% | 10 | Terminal-Bench 2.0; agent: ForgeCode | Official leaderboard via Epoch AI | Mar 2, 2026 |
| Terminal-Bench 2.0 | 74.8% | 18 | Terminal-Bench 2.0; agent: Terminus-KIRA | Official leaderboard via Epoch AI | Feb 19, 2026 |
| Terminal-Bench 2.0 | 61.4% | 45 | Terminal-Bench 2.0; agent: Gemini CLI | Official leaderboard via Epoch AI | Feb 19, 2026 |
FAQ
Where do these Gemini 3.1 Pro scores come from?
From Epoch AI Benchmarking Hub, LMArena leaderboard dataset, MMLU-Pro leaderboard (TIGER-Lab), and τ²-bench leaderboard submissions. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.