According to this thread, the overhead over pure WSGI seems low: https://groups.google.com/forum/#!topic/python-tulip/AgKcuC-v7aY https://groups.google.com/for

According to this thread, the overhead over pure WSGI seems low: https://groups.google.com/forum/#!topic/python-tulip/AgKcuC-v7aY https://groups.google.com/forum/#!topic/python-tulip/AgKcuC-... I'm wondering why the framework benchmark results are so different, even though they should include the same code: http://www.techempower.com/benchmarks/#section=data-r10&hw=peak&test=json&l=1kw http://www.techempower.com/benchmarks/#section=data-r10&hw=p...