Where to find the exact code that was used for Round22?
The ntex[sailfish] was the fastest in the single query category.
But there is no test with this display_name on the R22 tag nor on main
The other weird thing is that sailfish is not present in the implementation as a dependency
My understanding was that the benchmark_config.json display_name property is what's shown in the charts. It's true most of the times, but in some cases it's not.
There is also the go fiber-prefork. I have found the code that's likely what was used, but couldn't match it to what's shown in the chart.
That tag is correct. That is the right code and the numbers match up to: https://tfb-status.techempower.com/results/66d86090-b6d0-46b3-9752-5aa4913b2e33 but there does seem to be an issue with a couple of display names, probably due to some legacy code in how we attribute display names. Will take a look when I have some more time. Thanks for bringing it to our attention.