Gemini 3.8 Flash: benchmark results
Provider: Google. Released 2026-09-02. Access: API.
Unified ELO 1778 ± 1, rank #6 of 1392 rated models, from 37 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BioMysteryBench Human-Difficult | 56.5 | Accuracy (self-reported) | 100 |
| LLM Stats (Terminal-Bench 2.1) | 89.4 | Score (%) | 100 |
| LVBench | 87.8 | Score (self-reported) | 100 |
| AI Chess Leaderboard (Continuation) | 1772 | Elo | 99.2 |
| AI Chess Leaderboard (Reasoning) | 1849 | Elo | 99.1 |
| Design Arena (UI Components) | 1368 | Elo | 99 |
| BenchmarkList ECI | 152.3 | Capability Index (ECI) | 98.4 |
| SimpleBench | 82.4 | Score (AVG@5) | 98 |
| BenchLM | 78.4 | Overall Score | 97.5 |
| SvelteBench | 100 | Average pass@1 (%) | 96.9 |
| LLM Stats Score | 50.97 | LLM Stats Score (conservative rating) | 96 |
| Design Arena (Game Dev) | 1344 | Elo | 95 |
Interactive version: theaggregate.ai/model?slug=gemini-3-8-flash · How It Works · Data refreshed daily, snapshot 2026-09-05.