Spaces:
Running
Running
Grok 4.20
#1
by shino256 - opened
Will Grok 4.20 be included in benchmark results soon? Only Grok 4.1 is on the list, but Grok 4.20 has been released recently.
I evaluated the early versions of 4.20 and the results were poor, so I suspected there was an issue with the Grok API and did not report those results on the leaderboard. I will check again soon.
The current version of Grok 4.20 is still bad. Apparently, the model suffered a significant drop in quality for Polish.
sdadas changed discussion status to closed