Look at it this way the leaked benchmarks are like those highlight reels for athletes. Sure, the sprinter looks amazing in a 100 meter dash… but you don’t know how they’ll do over 10 km. The leaks show “best‑case scenarios.” In everyday workflows compile times, many browser tabs, VMs the gains might be more incremental. Until a full review or public benchmark suite shows consistent results, treat the leak numbers as marketing material, not gospel.