Gemini 3 Flash (Minimal): benchmark results
Provider: Google. Released 2025-12-17. Access: API.
Aggregate Rating 1721 ± 15, rank #303 of 2170 rated models, from 74 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BFI-Adapt | 34.8 | BFI-Adapt composite (%; mean over the 27 event-trait pairs w | 100 |
| DocFormBench (DocFormFlow) | 80.36 | Format accuracy (%) on DocFormBench: share of required forma | 100 |
| DocFormBench (MCP-Msoffice) | 22.38 | Format accuracy (%) on DocFormBench: share of required forma | 100 |
| FinBalance - Replayed Balance Sheet | 54.8 | Replayed balance-sheet accuracy (%): share of records whose | 100 |
| GR-Ben - Analogical Reasoning | 60.8 | F1 (%) of first-error identification: the harmonic mean of t | 100 |
| GR-Ben - Biology | 57.6 | F1 (%) of first-error identification: the harmonic mean of t | 100 |
| GR-Ben - Computer Science | 61.6 | F1 (%) of first-error identification: the harmonic mean of t | 100 |
| GR-Ben - Deductive Reasoning | 46.3 | F1 (%) of first-error identification: the harmonic mean of t | 100 |
| GR-Ben - Inductive Reasoning | 74.3 | F1 (%) of first-error identification: the harmonic mean of t | 100 |
| GR-Ben - Mixed-Form Reasoning | 67.4 | F1 (%) of first-error identification: the harmonic mean of t | 100 |
| LexKairos - Action Limitation Reasoning | 83.02 | F1 score (%; 109 items: decide whether a limitation period f | 100 |
| LexKairos - Effective Date Inquiry | 87.84 | Accuracy (%; 485 items: generate the effective date of a nam | 100 |
Interactive version: theaggregate.ai/model?slug=gemini-3-flash-minimal · How It Works · Data refreshed daily, snapshot 2026-11-05.