Gemini 2.5 Pro (Preview 06-05): benchmark results
June 5 Gemini 2.5 Pro preview snapshot. Provider: Google. Released 2025-06-05. Access: API.
Unified ELO 1684 ± 1, rank #78 of 1392 rated models, from 41 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Stats (Vibe-Eval) | 67.2 | Score (%) | 100 |
| SEAL - VISTA | 54.63 | Score | 98.4 |
| SEAL - Fortress | 61.68 | Score | 96.4 |
| Web-Bench | 25.3 | Pass@1 (%) | 95.7 |
| Balrog | 43.3 | Score (self-reported) | 93.3 |
| SEAL - TutorBench | 55.65 | Score | 92.3 |
| SpeechMap Compliance | 89 | % Requests Completed | 89.4 |
| CadEval | 64 | Score (self-reported) | 88.9 |
| Context Arena MRCR (2-needle) | 77.5 | AUC@1M (%) | 88.8 |
| Wolfram LLM Benchmarking Project | 59.1 | Correct Functionality (%) | 87.3 |
| Fiction.LiveBench | 87.5 | Accuracy (%) | 87.2 |
| LiveCodeBench | 84.3 | Pass@1 avg (%) | 85.2 |
Interactive version: theaggregate.ai/model?slug=gemini-2-5-pro-preview-06-05 · How It Works · Data refreshed daily, snapshot 2026-09-05.