Grok 4.20 0309 (Non-reasoning) — benchmark results
March 9, 2026 Grok 4.20 snapshot evaluated with reasoning disabled. Provider: xAI. Released 2026-03-09. Access: API.
Unified ELO 1639 ± 19, rank #393 of 1776 rated models, from 59 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI Leaderboard | 51.77 | UGI Score | 92.9 |
| UGI - Writing | 47.15 | Writing Score | 87.8 |
| UGI - Natural Intelligence | 40.76 | NatInt Score | 87.6 |
| AA Humanity's Last Exam | 22.52 | Accuracy (%) | 82.6 |
| AA Omniscience - Software Engineering (SWE) - Java | 31 | Accuracy (%) | 80.7 |
| AA Omniscience - Software Engineering (SWE) - Python | 35.5 | Accuracy (%) | 77 |
| AA Omniscience - Law | 20.9 | Accuracy (%) | 76.3 |
| AA Omniscience - Software Engineering (SWE) - PHP | 38 | Accuracy (%) | 72.8 |
| AA Omniscience - Humanities & Social Sciences | 26.9 | Accuracy (%) | 72.3 |
| AA Omniscience - Health | 25 | Accuracy (%) | 72.1 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 28 | Accuracy (%) | 71.9 |
| AA Omniscience - Software Engineering (SWE) | 34.7 | Accuracy (%) | 71 |
Interactive version: theaggregate.ai/model?slug=grok-4-20-0309-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.