Grok 4.20 0309 v2 (Reasoning): benchmark results
Provider: xAI. Released 2026-03-09. Access: API.
Unified ELO 1648 ± 1, rank #250 of 1761 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA IFBench | 81.22 | Accuracy (%) | 98.9 |
| AA GPQA Diamond | 91.11 | Accuracy (%) | 91.7 |
| AA Omniscience | 14.83 | Score | 89.8 |
| AA TAU-2 Bench | 92.98 | Accuracy (%) | 89.1 |
| AA Humanity's Last Exam | 34.52 | Accuracy (%) | 86.4 |
| AA Terminal-Bench Hard | 37.88 | Accuracy (%) | 84.9 |
| AA CritPt | 6.57 | Accuracy (%) | 81.4 |
| Artificial Analysis Intelligence Index | 30.02 | Intelligence Index | 80.9 |
| AA Omniscience - Software Engineering (SWE) | 42.9 | Accuracy (%) | 74.5 |
| AA Omniscience - Science, Engineering & Mathematics | 35.6 | Accuracy (%) | 71.4 |
| AA-Omniscience Accuracy | 27.48 | Accuracy (%) | 69.4 |
| AA MMMU-Pro | 74.57 | Accuracy (%) | 68 |
Interactive version: theaggregate.ai/model?slug=grok-4-20-0309-v2-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.