Grok 4.20 is an AI model developed by xAI (United States), first published in February 2026. It works in the language domain, on tasks such as language modeling/generation.
Epoch AI has no training-compute estimate for this model. The model has 500,000,000,000 parameters.
Access: API access. Its weights are not openly released. Epoch AI rates the confidence of this record as likely.
| 01 | Vending-Bench 2 | Score | 4662.8 |
| 02 | WebDev Arena | Arena Score | 1395.2 |
| 03 | ALE Bench | Performance | 1150.3 |
| 04 | Epoch Capabilities Index | ECI Score | 152.6 |
| 05 | ForecastBench | Overall score | 61.4 |
| 06 | ARC-AGI | Score | 0.9 |
| 07 | ARC-AGI-2 | Score | 0.65 |
| 08 | Terminal-Bench | Accuracy mean | 0.57 |
| 09 | WeirdML | Accuracy | 0.52 |
| 10 | CL Bench | Overall | 0.22 |
| 11 | CL Bench Life | Overall | 0.12 |
| 12 | Blueprint Bench 2 | Score | 0 |