The major contribution was given by @fantix, I appreciate him for the hard work.

However, I'm deeply disappointed by the low interest on Python 3 issues in the gevent team. @fantix's Python 3 branch was there for more than a year. I've used it on production machines since last year, and it has worked great. However, the actual merging didn't happen until weeks ago, so I couldn't help but think they were not interested. Thankfully it finally happened, but it took too much time.

This is going to cast some big ripples. There are a lot of packages that have been waiting for this. It might have been better in the long run if every one got on the AsynchIO train, but if nothing else the python2/3 saga has shown that standards can have a lot of momentum.

How does this relate to https://www.python.org/dev/peps/pep-0492/ ?

That's a big questions. 0492 was about bringing asyncIO in to python 3.5 . AsyncIO is part of the core python 3 where gevent is a package (origninally designed for python 2). They accomplish broadly the same thing. I think that AsycIO is a better longterm solution, but gevent has more momentum.

Thanks, with async and await being included in the language, will gevent use the newly incorporated features internally, or are they exclusive?

Gevent will probably never have a strong dependency on AsyncIO. Their disagreement is very deep and it's unlikely you could do things in a philosophically gevent way with the asyncIO tools.

also supports pypy

Thanks so much for this. I've been using gevent 1.1a2 to support Python 3.4. Does this mean we are out of Alpha? __init__.py shows 1.1a3 tho!