I think TechEmpower are doing a great job on these generally, but the plaintext benchmark is problematic because they allow pipelining. (Which isn’t enabled in browsers, isn’t typically used in request libraries and tools, and has generally been sidelined) Frameworks which are tailored with that in place can get a big boost in the ratings there.
If you want a measure of plain request/response I’d use the JSON benchmarks as the baseline.
No comments yet.