Gemini 3 Pro (High): benchmark results
Gemini 3 Pro evaluated at the high reasoning-effort setting. Provider: Google. Released 2025-11-18. Access: API.
Unified ELO 1635 ± 1, rank #298 of 1761 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MCPMark | 53.94 | Pass@1 (%) | 97.4 |
| VitaBench | 31.5 | Cross-Scenario Avg@4 (%) | 95 |
| Pencil Puzzle Bench - Light Up | 20 | Direct-ask Success Rate (%) | 89 |
| Pencil Puzzle Bench - Slitherlink | 13.3 | Direct-ask Success Rate (%) | 88 |
| Pencil Puzzle Bench - Nurimisaki | 6.7 | Direct-ask Success Rate (%) | 71 |
| Pencil Puzzle Bench - Shikaku | 6.7 | Direct-ask Success Rate (%) | 71 |
| GISA | 15.28 | Overall Score | 66.7 |
| Pencil Puzzle Bench - Norinori | 6.7 | Direct-ask Success Rate (%) | 63 |
| Pencil Puzzle Bench - Hitori | 6.7 | Direct-ask Success Rate (%) | 56 |
| APEX-Agents | 34.1 | Mean Score (ReAct) (self-reported) | 52.3 |
| Pencil Puzzle Bench - Heyawake | 0 | Direct-ask Success Rate (%) | 50 |
| Pencil Puzzle Bench - Sashigane | 0 | Direct-ask Success Rate (%) | 50 |
Interactive version: theaggregate.ai/model?slug=gemini-3-pro-high · How It Works · Data refreshed daily, snapshot 2026-09-05.