Agnes 2.5 Pro Beta: benchmark results
Provider: Other. Released 2026-08-26. Access: API.
Aggregate Rating 1698 ± 38, rank #365 of 2168 rated models, from 34 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Long Context Reasoning | 83 | Accuracy (%) | 95.1 |
| AA-Omniscience Hallucination Rate | 32.8 | Hallucination Rate (%) | 90.5 |
| AA GPQA Diamond | 90.51 | Accuracy (%) | 90.3 |
| Artificial Analysis Intelligence Index | 35.24 | Intelligence Index | 87.5 |
| AA CritPt | 15.71 | Accuracy (%) | 84.9 |
| AA Humanity's Last Exam | 37.53 | Accuracy (%) | 84.5 |
| Tau3 Banking | 35.67 | Success Rate (%) | 78.4 |
| AA-Omniscience Index - Science, Engineering & Mathematics | 2 | Omniscience Index | 77.2 |
| AA-Omniscience Index - Humanities & Social Sciences | -0.5 | Omniscience Index | 75.8 |
| AA-Omniscience Index - Software Engineering (SWE) - TypeScript | 12.22 | Omniscience Index | 69.9 |
| AA-Omniscience Index - Law | -10.1 | Omniscience Index | 69.5 |
| AA Terminal-Bench 2.1 | 69.66 | Pass@1 (%) | 68.9 |
Interactive version: theaggregate.ai/model?slug=agnes-2-5-pro-beta · How It Works · Data refreshed daily, snapshot 2026-10-20.