Gemini 3.8 Flash (Medium): benchmark results
Provider: Google. Released 2026-09-02. Access: API.
Aggregate Rating 1852 ± 12, rank #66 of 2168 rated models, from 112 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Wolfram LLM Benchmarking Project | 73.8 | Correct Functionality (%) | 99.2 |
| AA Long Context Reasoning | 84 | Accuracy (%) | 98 |
| AA GPQA Diamond | 93.54 | Accuracy (%) | 97.4 |
| cua-speedrun - OSWorld (50 Tasks) | 91.62 | Task success (%), one run per task in the cua-speedrun Modal | 96.9 |
| AA-Omniscience Index - Law | 40 | Omniscience Index | 96.4 |
| LLM Chess (Saplin) | 1526.51 | Native anchored-opponent Elo (signed points) | 94.4 |
| ARC-AGI-1 | 97.5 | Accuracy (%) | 94.3 |
| AA GDP.pdf - Insurance - Criterion Pass Rate | 86.19 | Criterion Pass Rate (%) | 94.2 |
| AA-Omniscience Index - Humanities & Social Sciences | 29.5 | Omniscience Index | 94.1 |
| AA-Omniscience Index - Software Engineering (SWE) - PHP | 70 | Omniscience Index | 94 |
| Tau3 Banking | 45.77 | Success Rate (%) | 94 |
| AA-Omniscience Index - Business | 22.5 | Omniscience Index | 93.9 |
Interactive version: theaggregate.ai/model?slug=gemini-3-8-flash-medium · How It Works · Data refreshed daily, snapshot 2026-10-20.