Model card
Gemini 3.1 Pro.
Googleproprietary1 current SOTA
Google Gemini 3.1 Pro canonical model row.
§ 02 · Benchmarks
Every benchmark Gemini 3.1 Pro has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | LiveCodeBench Pro | Computer Code · Code Generation | elo | 2887.00 | #1 | — | source ↗ |
| 02 | HLE | Reasoning · Multi-step Reasoning | accuracy | 46.4% | #3 | — | source ↗ |
| 03 | SWE-Bench Verified | Computer Code · Code Generation | resolve-rate | 80.6% | #4 | 2026-02-19 | source ↗ |
| 04 | SWE-bench Verified | Agentic AI · SWE-bench | resolve-rate | 80.6% | #4 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where Gemini 3.1 Pro actually performs.
§ 05 · Related models
Other Google models scored on Codesota.
§ 06 · Sources & freshness
Where these numbers come from.
livecodebench-pro-official
1
result
scale-hle-official
1
result
google-blog
1
result
editorial
1
result
4 of 4 rows marked verified.