Gemini 3.1 Flash Lite (Preview) (High) — benchmark results
Gemini 3.1 Flash Lite (Preview) evaluated at the high reasoning-effort setting. Provider: Google. Released 2026-05-07. Access: API.
Unified ELO 1692 ± 27, rank #288 of 1776 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MedCode | 47.6 | Score (self-reported) | 71.7 |
| WeirdML | 52.19 | Average Score | 67.2 |
| ALE-Bench | 797.73 | Performance (Self-Refine x1) (self-reported) | 60.3 |
| APEX-Agents | 25 | Mean Score (ReAct) (self-reported) | 35.9 |
| Vals AI CaseLaw v2 | 54.98 | Accuracy (%) | 35.1 |
| Finance Agent v1.1 | 46.12 | Score (self-reported) | 30.9 |
| Vals Multimodal Index | 41.35 | Score (self-reported) | 15.8 |
| Finance Agent v2 | 29.99 | Score (self-reported) | 10 |
| MedScribe | 63.9 | Score (self-reported) | 7.7 |
| Vals Index | 36.2 | Score (self-reported) | 4.2 |
Interactive version: theaggregate.ai/model?slug=gemini-3-1-flash-lite-preview-high · How the rankings work · Data refreshed daily, snapshot 2026-07-22.