Gemini 3.1 Pro (Preview) (Low): benchmark results
Gemini 3.1 Pro (Preview) evaluated at the low reasoning-effort setting. Provider: Google. Released 2026-02-19. Access: API.
Aggregate Rating 1740 ± 11, rank #260 of 2168 rated models, from 110 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ChessImageBench | 20.8 | All Correct (%) | 100 |
| FinLifeBench - Life-Event History - Event-Anchor F1 | 74.8 | F1 (%; event type with first-establishing session) | 100 |
| FinLifeBench - Life-Event History - Exact History Match | 11.3 | Checkpoints with the exact history (%) | 100 |
| TeachObs - Scene Coding (Transcript + Frame) | 51 | Macro F1 (%; 39 observation codes per 15-second classroom sc | 100 |
| VSTAT | 44.4 | Average score (%) over all questions: 1,500 questions on 834 | 100 |
| VSTAT - Attribute | 54.1 | Score (%) on questions about attribute-type state elements: | 100 |
| VSTAT - Count | 42.6 | Score (%) on questions about count-type state elements: 1,50 | 100 |
| VSTAT - Set | 51.9 | Score (%) on questions about set state structures: 1,500 que | 100 |
| Japanese G2P (Parse Mode) | 0.62 | Kana character error rate (%; can exceed 100) in parse mode | 97.2 |
| DGEval - IMDG Code MCQ | 94.4 | Accuracy (%) | 97.1 |
| VSTAT - Atomic | 39.5 | Score (%) on questions about atomic state structures: 1,500 | 95.5 |
| VSTAT - Dictionary | 38.7 | Score (%) on questions about dictionary (entity-to-value) st | 95.5 |
Interactive version: theaggregate.ai/model?slug=gemini-3-1-pro-preview-low · How It Works · Data refreshed daily, snapshot 2026-10-20.