The linked post points to OEIS A014233[1] for establishing their set of Miller-Rabin[2] bases, though it's actually possible to find smaller sets.<p>I remember asking about this on StackExchange some years ago [3], which pointed me to Wojciech Izykowski's site[4], on which "best known" base sets are tracked. For example, instead of considering the four bases {2,3,5,7} to cover all 32-bit integers, it would suffice to consider the three integers {4230279247111683200, 14694767155120705706, 16641139526367750375}.<p>This becomes more interesting the higher the bound you seek --- for example, instead of checking the first 11 prime bases for 64-bit integers, you only need to check the seven bases: 2, 325, 9375, 28178, 450775, 9780504, 1795265022.<p>[1]: <a href="https://oeis.org/A014233" rel="nofollow">https://oeis.org/A014233</a><p>[2]: <a href="https://en.wikipedia.org/wiki/Miller%E2%80%93Rabin_primality_test" rel="nofollow">https://en.wikipedia.org/wiki/Miller%E2%80%93Rabin_primality...</a><p>[3]: <a href="https://math.stackexchange.com/questions/1004807/" rel="nofollow">https://math.stackexchange.com/questions/1004807/</a><p>[4]: <a href="https://miller-rabin.appspot.com" rel="nofollow">https://miller-rabin.appspot.com</a> or <a href="https://web.archive.org/web/20260225175716/https://miller-rabin.appspot.com/" rel="nofollow">https://web.archive.org/web/20260225175716/https://miller-ra...</a> if hugged to death