Depends on who made the mold and how careful you cast. So I'll start my answer kindly. Try sorting a lot of bullets by weight, shoot the different lots, in what ever weight variations you want to test. If you can't see a difference in accuracy with the different lots, don't ever do it again. --- Now I'm going to try to discourage everyone from weigh checking.
In my book I go into some depth on this issure, and give a test I performed the last time I weigh checked bullets, 20 some years ago! Since that test I like to say. "Weigh sorting is something people do when they don't know how to make their bullets shoot accurately." I expect that is a bit candid, perhaps even mean, but I don't like to have people lead into chasing their tails with meaningless work that ruins all the fun of casting bullets. So the statement is kinder than reccomending weigh sorting.
To keep the account real brief, unsorted shot as well or better than straight run, and the culls, by weight, shot as well as the close weight groups.
If you are using an LBT mold and cast per the instructions there will be no weight variation in bullets from each cavity, though there is often 1 or 2 tenths of a grain difference in cavities. I once cast 100 bullets and weigh checked each one. The result was exactly 50 of each weight.
The only reason weight variations in cast bullets cause inaccuracy is if they have invisible internal voids, and LBT mold design eliminates them if you run it hot and keep the mold tight shut with every cast. Bullet design has far more to do with good accuracy than slight weight variations, so if you are straining out nats to tighten accuracy with a poor design, you are fighting a losing battle. If the design is strong and the bullet properly fitted, and barrel lapped (with most barrels) slight weight variations won't effect accuracy enough for the average shooter to find it. Especially handgunners.