Gemini 2.0 Pro: benchmark results
Provider: Google. Released 2025-02-05. Access: API.
Aggregate Rating 1639 ± 14, rank #697 of 10916 rated models, from 289 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AesGuide | 1.19 | Mean of the completeness, preciseness and relevance scores o | 100 |
| AesGuide - Completeness | 1.09 | Completeness score (coverage of the reference critique) on A | 100 |
| AesGuide - Expert Assessment | 1.16 | Expert assessment score on 100 sampled AesGuide photos: ten | 100 |
| AesGuide - Preciseness | 1.12 | Preciseness score (no inaccurate or conflicting details) on | 100 |
| AesGuide - Relevance | 1.36 | Relevance score (on-topic, actionable guidance) on AesGuide' | 100 |
| ECLeKTic | 41.6 | Score (imported) | 100 |
| Open LMM Reasoning - DynaMath - Subject-plane geometry | 50.6 | Accuracy (%) | 100 |
| Open LMM Reasoning - DynaMath - Subject-solid geometry | 46.7 | Accuracy (%) | 100 |
| Open LMM Reasoning - DynaMath - Subject-solid geometry; Avg | 68.7 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVerse | 67.3 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVerse - Analytic | 79.1 | Accuracy (%) | 100 |
| Open LMM Reasoning - MathVerse - Applied | 69.7 | Accuracy (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gemini-2-0-pro · How It Works · Data refreshed daily, snapshot 2026-11-08.