Gemini 3 Pro — benchmark results
Google's Pro-tier Gemini 3 model for advanced reasoning and general tasks. Provider: Google. Released 2025-11-18. Access: API.
Unified ELO 1847 ± 13, rank #105 of 1776 rated models, from 223 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BALROG MiniHack (LLM) | 40 | Progress (%) | 100 |
| BALROG NetHack (LLM) | 6.8 | Progress (%) | 100 |
| Diplomacy: Overall Performance | 60.3 | Score | 100 |
| EmoBench-M | 70.5 | Average Score | 100 |
| JMMMU-Pro - Culture Agnostic | 80.42 | Accuracy (%) | 100 |
| JMMMU-Pro - Culture Specific | 95 | Accuracy (%) | 100 |
| JMMMU-Pro - Japanese Art | 91.33 | Accuracy (%) | 100 |
| JMMMU-Pro - Japanese Heritage | 96.67 | Accuracy (%) | 100 |
| JMMMU-Pro - Japanese History | 95.33 | Accuracy (%) | 100 |
| JMMMU-Pro - Overall | 87.05 | Accuracy (%) | 100 |
| JMMMU-Pro - World History | 96.67 | Accuracy (%) | 100 |
| LLM Stats (VideoMMMU) | 87.6 | Score (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gemini-3-pro · How the rankings work · Data refreshed daily, snapshot 2026-07-22.