Grok 4.3 (Low): benchmark results
Provider: xAI. Released 2026-04-30. Access: API.
Aggregate Rating 1713 ± 30, rank #320 of 2168 rated models, from 41 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| TeachObs - Lesson Narrative Coverage | 33.1 | Reference coverage (%; share of the experts' atomic claims a | 100 |
| AA IFBench | 80.95 | Accuracy (%) | 98.7 |
| AA-Omniscience Hallucination Rate | 17.39 | Hallucination Rate (%) | 98.6 |
| AA-Omniscience Index - Health | 7.4 | Omniscience Index | 90.3 |
| AA-Omniscience Index - Business | 11.8 | Omniscience Index | 88.7 |
| AA-Omniscience Index - Humanities & Social Sciences | 15.9 | Omniscience Index | 86.7 |
| AA Omniscience | 13.93 | Score | 85.3 |
| AA-Omniscience Index - Law | 10 | Omniscience Index | 84.9 |
| AA-Omniscience Index - Science, Engineering & Mathematics | 11.3 | Omniscience Index | 84.9 |
| AA-Omniscience Index - Software Engineering (SWE) - Java | 10 | Omniscience Index | 83.9 |
| AA TAU-2 Bench | 88.89 | Accuracy (%) | 83.3 |
| Tau2-Bench Telecom | 88.89 | Success Rate (%) | 83.1 |
Interactive version: theaggregate.ai/model?slug=grok-4-3-low · How It Works · Data refreshed daily, snapshot 2026-10-20.