The above chart compares the acceleration factor achieved by different compilers/implementations when accelerating Python code on selected benchmarks. The acceleration factor is calculated as the time it took the Python interpreter to run the benchmark divided by the time it took the compiler/implementation being analysed to run the benchmark, so higher is better.
Specifications of the benchmarking machine:
More benchmarks are coming.
The benchmarks are strongly inspired from benchmarks that can be found in the Pybenchmarks and Python Performance Benchmark suites. We plan to use the benchmarks from these websites in the near future. The benchmarks were only run once and we are in the process of establishing a more rigorous benchmarking process. It was done to get a broad idea of the performance gain that Flyable can bring and there is still a lot of improvements to be made.