Grok 4.3 (High): benchmark results
Grok 4.3 evaluated at the high reasoning-effort setting. Provider: xAI. Released 2026-04-30. Access: API.
Aggregate Rating 1666 ± 11, rank #471 of 2168 rated models, from 133 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA IFBench | 81.29 | Accuracy (%) | 99.1 |
| Tau2-Bench Telecom | 97.66 | Success Rate (%) | 98.1 |
| AA TAU-2 Bench | 97.66 | Accuracy (%) | 97.9 |
| AA-Omniscience Hallucination Rate | 25.76 | Hallucination Rate (%) | 95.3 |
| AA-Omniscience Index - Health | 9.41 | Omniscience Index | 91.9 |
| MLCR-AA - Conciseness | 98.89 | Conciseness (%) | 90.5 |
| AA-Omniscience Index - Science, Engineering & Mathematics | 17.33 | Omniscience Index | 90.3 |
| AA GPQA Diamond | 90.1 | Accuracy (%) | 89.6 |
| AA-Omniscience Index - Business | 13.1 | Omniscience Index | 89.2 |
| AA-Omniscience Index - Law | 17.7 | Omniscience Index | 89.1 |
| AA-Omniscience Index - Humanities & Social Sciences | 18.32 | Omniscience Index | 88.9 |
| AA Omniscience | 17.98 | Score | 87.2 |
Interactive version: theaggregate.ai/model?slug=grok-4-3-high · How It Works · Data refreshed daily, snapshot 2026-10-20.