Grok 4.1 (Thinking): benchmark results
Provider: xAI. Released 2025-11-17. Access: API.
Aggregate Rating 1657 ± 25, rank #500 of 2170 rated models, from 67 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| URSA Single-Step Retrosynthesis | 0.49 | Average per-target ChemCensor score (0 to 5) over the model' | 100 |
| Chatbot Arena (Text - Medicine & Healthcare) | 1500 | Arena Score | 95.3 |
| Chatbot Arena (Text - Polish) | 1486 | Arena Score | 90.9 |
| URSA Single-Step Retrosynthesis - Max ChemCensor | 1.75 | Average over the 100 URSA-expert-2026 novel targets of the h | 89.5 |
| Chatbot Arena (Text - Legal & Government) | 1475 | Arena Score | 88.1 |
| Chatbot Arena (Text - English) | 1473 | Arena Score | 87.8 |
| Chatbot Arena (Text - Non-English) | 1453 | Arena Score | 87.5 |
| Chatbot Arena (Text) | 1465 | Elo | 87.3 |
| Chatbot Arena (Text - Spanish) | 1466 | Arena Score | 87 |
| Chatbot Arena (Text - Russian) | 1462 | Arena Score | 85.7 |
| Chatbot Arena (Text - Business, Management & Financial Ops) | 1459 | Arena Score | 84.9 |
| Chatbot Arena (Text - Software & IT Services) | 1496 | Arena Score | 84.9 |
Interactive version: theaggregate.ai/model?slug=grok-4-1-thinking · How It Works · Data refreshed daily, snapshot 2026-11-05.