That claim keeps contradicted hard by other parties, who say Mythos beats 5.5 resoundingly on both autonomous search and discovery and creation of complex explo

That claim keeps contradicted hard by other parties, who say Mythos beats 5.5 resoundingly on both autonomous search and discovery and creation of complex exploit chains. There might be a harness difference, but also, this CTF-type benchmark might not capture the capability difference fully.