Day-one estimate
Calculated from an estimated parameter count — the maker hasn't published one. Every model on this site is calculated the
same way, so the comparison holds even where the absolute value carries uncertainty.
Our figures are good to roughly a factor of two to three, which is why we only ever
state a difference of 5× or more.
We cannot pin this one down. xAI has never published how Grok 4.20 is built, and the two plausible answers are far apart — which is why a range is shown rather than a figure. It is listed because leaving it out would reward the silence, and it is barred from every headline claim on this site.
Is there a better choice?
A lighter model that does the job
Nothing we track is both at least as capable as Grok 4.20 and meaningfully lighter. On our figures this is already a sensible choice for its standard of work.
What xAI publishes
Energy cannot be worked out without knowing how big a model is. Of the 8 models
from xAI that we track, 0 can be assessed at all — the rest publish
too little for anyone outside the company to calculate anything.