Mythos cracked this, mythos cracked that. But have they actually attempted to do the same with Opus?

I am skeptical about the alleged super capabilities of mythos/fable. I do believe it's an upgrade over Opus, but is it really that much of an upgrade? I mean sure there have been reports of mythos finding vulnerabilities in many places which I assume is legit, but I would like to know whether this has been properly AB tested? Like did they attempt to find the vulnerabilities using the same prompting techniques with Opus and got significantly weaker results? That's something that would convince me it really