Grok 4.20 Multi-Agent: benchmark results
xAI Grok 4.20 evaluated in multi-agent mode. Provider: xAI. Released 2026-03-09. Access: API.
Aggregate Rating 1754 ± 19, rank #113 of 1656 rated models, from 69 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI Leaderboard | 70 | UGI Score | 100 |
| ERR-EVAL - Trick Questions | 6.2 | Track score | 95.8 |
| UGI - Writing | 63.13 | Writing Score | 93.1 |
| Bullshit Benchmark | 67.3 | BS Detection Rate (%) | 92.3 |
| BenchTable - STEM | 79.4 | Weighted Score (%) | 90.7 |
| UGI - Natural Intelligence | 56.34 | NatInt Score | 90.6 |
| Chatbot Arena (Text - Russian) | 1478 | Arena Score | 90.5 |
| Chatbot Arena (Text - Non-English) | 1460 | Arena Score | 90 |
| Chatbot Arena (Text - Creative Writing) | 1447 | Arena Score | 89.3 |
| BenchTable | 72.4 | Total Score (%) | 89 |
| Chatbot Arena (Text) | 1470 | Elo | 89 |
| Chatbot Arena (Text - German) | 1474 | Arena Score | 88.1 |
Interactive version: theaggregate.ai/model?slug=grok-4-20-multi-agent · How It Works · Data refreshed daily, snapshot 2026-10-20.