Grok 4.5 (High): benchmark results
Grok 4.5 evaluated at the high reasoning-effort setting. Provider: xAI. Released 2026-07-08. Access: API.
Unified ELO 1695 ± 1, rank #119 of 1761 rated models, from 80 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Vals AI SkillsBench | 66.03 | Accuracy (%) | 100 |
| SWE-rebench | 63.78 | Resolved (%) | 99 |
| Epoch AI - Dtbench | 96.53 | Score | 97.5 |
| AA GPQA Diamond | 93.13 | Accuracy (%) | 96.1 |
| AA Omniscience - Health | 48.4 | Accuracy (%) | 95.1 |
| AA Omniscience - Science, Engineering & Mathematics | 52.32 | Accuracy (%) | 95.1 |
| AA Omniscience | 25.32 | Score | 94.3 |
| AA Humanity's Last Exam | 42.68 | Accuracy (%) | 94.1 |
| Artificial Analysis Intelligence Index | 45.48 | Intelligence Index | 94.1 |
| OTIS Mock AIME 2024-25 | 97.78 | Accuracy (%) | 93.6 |
| Vals AI Harvey Legal Agent Bench | 12.92 | Accuracy (%) | 93 |
| AA-Omniscience Accuracy | 51.55 | Accuracy (%) | 92.8 |
Interactive version: theaggregate.ai/model?slug=grok-4-5-high · How It Works · Data refreshed daily, snapshot 2026-09-05.