Gemini 2.5 Flash (Preview) (Non-reasoning) — benchmark results
Provider: Google. Released 2025-06-17. Access: API.
Unified ELO 1573 ± 33, rank #581 of 1841 rated models, from 8 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA MATH-500 | 92.6 | Accuracy (%) | 71.6 |
| AA MMLU-Pro | 78.31 | Accuracy (%) | 59.9 |
| AA LiveCodeBench | 40.63 | Pass@1 (%) | 48.8 |
| Artificial Analysis Intelligence Index | 11.66 | Intelligence Index | 38.8 |
| AA GPQA Diamond | 59.39 | Accuracy (%) | 37.4 |
| AA MMMU-Pro | 61.97 | Accuracy (%) | 35.8 |
| AA Humanity's Last Exam | 4.96 | Accuracy (%) | 32.1 |
| Epoch AI - Scicode | 23.26 | Score | 25.3 |
Interactive version: theaggregate.ai/model?slug=gemini-2-5-flash-preview-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-07-25.