Where is the result?
Are these results fake?
You can run it on your machine using bash all.sh. You need to install poop, awk, bc, jq, bash, rustc and clang++. You can also run a single benchmark using bash bench.sh name.
Does it mean that some language is faster/better than the other?
Not really. You can always use a custom library instead of the standard one. If you consider the quality of the implementation and standardization process in your comparison, then it might mean something.
Why do you use clang instead of g++?
I want to focus on the difference between standard library implementations, not LLVM and GCC optimizations.