Agnes 2.5 Pro Beta: benchmark results
Provider: Other. Released 2026-08-26. Access: API.
Unified ELO 1662 ± 1, rank #207 of 1761 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Long Context Reasoning | 83 | Accuracy (%) | 97 |
| BenchmarkList ECI | 144.03 | Capability Index (ECI) | 92.2 |
| AA-LCR | 78 | Accuracy (self-reported) | 90.8 |
| AA GPQA Diamond | 90.51 | Accuracy (%) | 90.4 |
| Artificial Analysis Intelligence Index | 39.51 | Intelligence Index | 90.1 |
| AA CritPt | 15.71 | Accuracy (%) | 89.4 |
| AA Humanity's Last Exam | 37.53 | Accuracy (%) | 88.9 |
| Tau3 Banking | 35.67 | Success Rate (%) | 78.2 |
| AA GDPval | 1379.04 | ELO | 75.9 |
| AA MMMU-Pro | 75.43 | Accuracy (%) | 72.1 |
| AA Omniscience | -10.53 | Score | 72.1 |
| AA Omniscience - Software Engineering (SWE) | 30.8 | Accuracy (%) | 58.7 |
Interactive version: theaggregate.ai/model?slug=agnes-2-5-pro-beta · How It Works · Data refreshed daily, snapshot 2026-09-05.