Gemini 2.0 Flash (Preview) — benchmark results
Google preview Gemini 2.0 Flash model row. Provider: Google. Released 2024-12-11. Access: API.
Unified ELO 1704 ± 9, rank #263 of 1776 rated models, from 587 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Public Goods Game | 45.2 | Avg. Contribution (%) | 100 |
| MEGA-Bench Task - Adapted CVBench Depth | 100 | Task Score (%) | 100 |
| MEGA-Bench Task - Autorater Control | 92.9 | Task Score (%) | 100 |
| MEGA-Bench Task - Electricity Load Estimate Plot | 79.3 | Task Score (%) | 100 |
| MEGA-Bench Task - Google Street View Line Reasoning | 60 | Task Score (%) | 100 |
| MEGA-Bench Task - Knowledge Graph Understanding | 80 | Task Score (%) | 100 |
| MEGA-Bench Task - Vibe Eval Phrase | 75.7 | Task Score (%) | 100 |
| MEGA-Bench Task - Weather Info Retrieval | 85.7 | Task Score (%) | 100 |
| MEGA-Bench Task - 3D Fragments Understanding | 42.9 | Task Score (%) | 98.8 |
| MEGA-Bench Task - ASCII Art Understanding | 71.4 | Task Score (%) | 98.8 |
| MEGA-Bench Task - App Interactive Operations Excel | 85.7 | Task Score (%) | 98.8 |
| MEGA-Bench Task - Autorater Aesthetics | 100 | Task Score (%) | 98.8 |
Interactive version: theaggregate.ai/model?slug=gemini-2-0-flash-preview · How the rankings work · Data refreshed daily, snapshot 2026-07-22.