Gemini 3.6 Flash (Medium): benchmark results
Provider: Google. Released 2026-07-21. Access: API.
Aggregate Rating 1783 ± 19, rank #169 of 2168 rated models, from 33 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Omni-Persona - Image-to-Image | 38.5 | Calibrated accuracy (%): item-weighted accuracy over answera | 100 |
| Wolfram LLM Benchmarking Project | 70.9 | Correct Functionality (%) | 96.6 |
| Chess Bench LLM | 1916 | Lichess-Classical-calibrated Glicko-2 rating (native signed | 93.9 |
| FrontierFinance - Financial Data & Modeling | 56.23 | Rubric qualification rate (%, macro average over queries, th | 91.7 |
| LLM Chess (Saplin) | 921.33 | Native anchored-opponent Elo (signed points) | 81.4 |
| Context Arena | 69.91 | Average Score (%) | 80.8 |
| ParseBench - Tables | 89.47 | GTRM Score (GriTS + TableRecordMatch) | 80.4 |
| Omni-Persona | 35.7 | Calibrated accuracy (%): item-weighted accuracy over answera | 80 |
| ForecastBench | 65.9 | Overall Score (higher is better) | 77 |
| DSV-Mem | 23 | Accuracy (%), 1,000 stateful visual-memory questions, mean o | 76.9 |
| FrontierFinance - Screening & Discovery | 36.11 | Rubric qualification rate (%, macro average over queries, th | 75 |
| ParseBench - Visual Grounding | 67.27 | Element Pass Rate | 71.8 |
Interactive version: theaggregate.ai/model?slug=gemini-3-6-flash-medium · How It Works · Data refreshed daily, snapshot 2026-10-20.