Gemini 3.1 Flash Lite (High): benchmark results

Gemini 3.1 Flash Lite evaluated at the high reasoning-effort setting. Provider: Google. Released 2026-05-07. Access: API.

Unified ELO 1642 ± 1, rank #379 of 3081 rated models, from 41 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Korean CSAT 2026 (Easy Mode) - Korean100Points (out of 100)92.5
SLR-Bench - Hard18Accuracy (%)88.2
SLR-Bench - Easy96Accuracy (%)87.3
SLR-Bench - Medium48Accuracy (%)87.3
ObviousBench97.92Answer pass³ (%)80.9
Korean CSAT 2026 (Easy Mode) - Chemistry I48Points (out of 50)79.9
Sonar LLM Leaderboard - Java - Pass Rate79.56Passing tests (%)69.4
Sonar LLM Leaderboard - Java - Bug Density0.73Bugs per 1,000 lines of code (lower is better)66.7
Korean CSAT 2026 (Easy Mode) - Total419.5Points (out of 450)64.6
Epoch AI - GPQA Diamond81.82Accuracy (%)64.1
OTIS Mock AIME 2024-2580Accuracy (%)63.1
Korean CSAT 2026 (Easy Mode) - Physics I35Points (out of 50)62.9

Interactive version: theaggregate.ai/model?slug=gemini-3-1-flash-lite-high · How It Works · Data refreshed daily, snapshot 2026-09-21.