A.X-K2: benchmark results
Provider: SK Telecom. Released 2026-07-29. Access: Open.
Unified ELO 1603 ± 1, rank #256 of 1392 rated models, from 63 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA CritPt | 8.86 | Accuracy (%) | 83.8 |
| AA Humanity's Last Exam | 29.56 | Accuracy (%) | 81.9 |
| AA GPQA Diamond | 85.66 | Accuracy (%) | 79.4 |
| Artificial Analysis Intelligence Index | 26.75 | Intelligence Index | 75.8 |
| BenchmarkList ECI | 130.29 | Capability Index (ECI) | 75.2 |
| AA Omniscience | -8.23 | Score | 74.9 |
| AA Long Context Reasoning | 70 | Accuracy (%) | 67.1 |
| AA Omniscience - Software Engineering (SWE) - R | 12 | Accuracy (%) | 62.1 |
| AA Omniscience - Software Engineering (SWE) - Python | 21.5 | Accuracy (%) | 61.8 |
| AA Omniscience - Science, Engineering & Mathematics | 30.7 | Accuracy (%) | 59.9 |
| Humanity's Last Exam (Self-Reported, No Tools) | 27.8 | Accuracy (%) | 58.8 |
| Nejumi 4 - ALT - Bias | 87.1 | Score (%) | 56.6 |
Interactive version: theaggregate.ai/model?slug=a-x-k2 · How It Works · Data refreshed daily, snapshot 2026-09-05.