Gemini 3.5 Flash (Minimal): benchmark results
Gemini 3.5 Flash evaluated at the minimal reasoning-effort setting. Provider: Google. Released 2026-05-19. Access: API.
Aggregate Rating 1720 ± 15, rank #303 of 2168 rated models, from 84 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Khondo (MonoRand) | 0.83 | Packet score (-0.5 to 1; mean of the clustering score, the a | 100 |
| Khondo (PolyInt) | 0.73 | Packet score (-0.5 to 1; mean of the clustering score, the a | 100 |
| Khondo (PolyRand) | 0.73 | Packet score (-0.5 to 1; mean of the clustering score, the a | 100 |
| UGI - Writing | 72.47 | Writing Score | 99 |
| UGI - Natural Intelligence | 72.31 | NatInt Score | 97.1 |
| SteerBench-Work - Pass^5 | 89.6 | Scenarios Correct in All 5 Trials (%) | 94.8 |
| AA Terminal-Bench Hard | 46.21 | Accuracy (%) | 92.9 |
| M-GATE - Grammatical Error Detection | 0.19 | Mean MCC over 30 languages (-1 to 1; temperature 0) | 92.6 |
| RideWay - Success Rate | 82.2 | Success rate (%; share of trials passing the three-judge LLM | 89.1 |
| MC-CXR - Misleading Prior Report | 21.7 | Accuracy under misleading context (%; the image plus a prior | 88.9 |
| Khondo (MonoSeq) | 0.9 | Packet score (-0.5 to 1; mean of the clustering score, the a | 87.5 |
| SteerBench-Work | 90.6 | Mean Trial Accuracy (%) | 86.2 |
Interactive version: theaggregate.ai/model?slug=gemini-3-5-flash-minimal · How It Works · Data refreshed daily, snapshot 2026-10-20.