Gemini 3.6 Flash (Low): benchmark results
Provider: Google. Released 2026-07-21. Access: API.
Aggregate Rating 1744 ± 35, rank #253 of 2168 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Chess Puzzles (Epoch AI) | 42 | Accuracy (%) | 76.6 |
| Epoch AI - GPQA Diamond | 86.36 | Accuracy (%) | 74.8 |
| LLM Chess (Saplin) | 498.4 | Native anchored-opponent Elo (signed points) | 69.5 |
| Context Arena | 63.04 | Average Score (%) | 68.4 |
| Epoch AI - Mystery Game Puzzles | 23 | Score | 63.4 |
| OTIS Mock AIME 2024-25 | 82.22 | Accuracy (%) | 63 |
| AutomationBench | 15.68 | Strict Task Pass Rate (%) | 62 |
| EnigmaForge | 43.7 | Task success (%) over 600 procedurally generated story insta | 58.3 |
| ARC-AGI-1 | 76.5 | Accuracy (%) | 54.8 |
| ARC-AGI-2 | 30.42 | Accuracy (%) | 53.5 |
| EnigmaForge - Intuition | 16.2 | Task success (%) on implicit-condition instances, where the | 33.3 |
| EnigmaForge - Fact F1 | 79.5 | World-fact recovery F1 (%) over 600 instances, all items | 29.2 |
Interactive version: theaggregate.ai/model?slug=gemini-3-6-flash-low · How It Works · Data refreshed daily, snapshot 2026-10-20.