"What gets us into trouble is not what we don't know, it's what we know for sure that just ain't so."
Please don't use this summary page to compare 2 interpreters -- use the direct comparison, for example -all benchmarks- Python 2 PyPy.
This box plot shows how many times slower, the fastest benchmark for interpreters were, compared to the fastest benchmarks in any of the interpreters.
Please don't obsess about which interpreter is shown 10th and which is shown 11th. You can see that the order would be different if it was based on the median scores instead of the geometric mean scores.
The direct comparison between 2 programming language implementations will give you a better understanding of the differences in program performance than this summary page.
In some cases, a median score shown here is faster but the median time ratio shown in the direct comparison is slower. In those cases, the range of measurements for those 2 programming language implementations is really quite similar.