xAI
Grok 4.5
4 benchmarks
Terminal-Bench 2.1
#12 of 27
83.3%
SWE-bench Pro
#5 of 43
64.7%
DeepSWE
#10 of 22
62.0%
DeepSWE
#10 of 22
53.0%
benchmark.space canonical public ledger of AI capability