Prisma vs JDBC: the benchmark that almost made me blame the wrong ORM

There's a discussion that surfaces every time someone posts an ORM benchmark: "of course JDBC is faster, you're measuring the abstraction". They're right, but only halfway. What nobody says is that the abstraction isn't the only culprit — sometimes the culprit is you, because you let an N+1 slip through without noticing.

I built . Editorial results are in results/comparison.csv and results/comparison.md.

What I'd like to know: in the stack you're using right now, do you have real visibility into the SQL/request count for each endpoint? Or do you assume the ORM handles it on its own?

This article was originally published on juanchi.dev.