Gemini 3.1 Flash Lite (Preview): benchmark results
Google preview Gemini 3.1 Flash Lite model for low-latency, high-volume workloads. Provider: Google. Released 2026-03-03. Access: API.
Unified ELO 1628 ± 1, rank #185 of 1392 rated models, from 410 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Dutch NLU - ScaLA NL | 77.57 | Linguistic acceptability Score (%) | 100 |
| EuroEval Estonian NLU - Estonian Valence | 67.41 | Sentiment classification Score (%) | 100 |
| EuroEval Ukrainian NLU - ScaLA UK | 52.39 | Linguistic acceptability Score (%) | 100 |
| EuroEval Danish NLU - ScaLA DA | 77.63 | Linguistic acceptability Score (%) | 99.8 |
| EuroEval Italian NLU - ScaLA IT | 79.29 | Linguistic acceptability Score (%) | 99.8 |
| EuroEval Finnish NLU - ScaLA FI | 59.08 | Linguistic acceptability Score (%) | 99.7 |
| EuroEval Icelandic NLU - Hotter and Colder Sentiment | 63.57 | Sentiment classification Score (%) | 99.7 |
| EuroEval Norwegian NLU - ScaLA NN | 73.39 | Linguistic acceptability Score (%) | 99.7 |
| EuroEval Bulgarian NLU - ScaLA BG | 83.09 | Linguistic acceptability Score (%) | 99.5 |
| EuroEval Croatian NLU - ScaLA HR | 76.62 | Linguistic acceptability Score (%) | 99.5 |
| EuroEval Danish NLU | 70.05 | NLU Average Score (%) | 99.5 |
| EuroEval Serbian NLU - ScaLA SR | 79.48 | Linguistic acceptability Score (%) | 99.5 |
Interactive version: theaggregate.ai/model?slug=gemini-3-1-flash-lite-preview · How It Works · Data refreshed daily, snapshot 2026-09-05.