Gemini 3.8 Flash (Medium): benchmark results
Provider: Google. Released 2026-09-02. Access: API.
Unified ELO 1735 ± 1, rank #47 of 1761 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Long Context Reasoning | 84 | Accuracy (%) | 98.7 |
| AA GPQA Diamond | 93.54 | Accuracy (%) | 97.3 |
| AA Omniscience - Law | 59.5 | Accuracy (%) | 96.7 |
| AA Omniscience | 28.58 | Score | 96.1 |
| Artificial Analysis Intelligence Index | 46.77 | Intelligence Index | 95.3 |
| AA MMMU-Pro | 84.22 | Accuracy (%) | 95 |
| Tau3 Banking | 45.77 | Success Rate (%) | 94.3 |
| AA Omniscience - Business | 44 | Accuracy (%) | 93.7 |
| AA Omniscience - Humanities & Social Sciences | 52.6 | Accuracy (%) | 93.6 |
| AA-Omniscience Accuracy | 52.98 | Accuracy (%) | 93.6 |
| AA Humanity's Last Exam | 42.12 | Accuracy (%) | 92.8 |
| Computer Anthology Terminal Tasks (Terminus-2) | 51.8 | pass@1 (%) | 92.3 |
Interactive version: theaggregate.ai/model?slug=gemini-3-8-flash-medium · How It Works · Data refreshed daily, snapshot 2026-09-05.