The Library of Congress and Twitter agreed back in April 2010 that the LoC could archive all public tweets from 2006 to April 2010. It's taken the Library of Congress a while to do that, almost three years in fact, and they have announced that they will complete the archive this month.
The archive now contains 170B tweets and is growing by 500M tweets per day. Compare that with the 140M tweets per day that it was archiving back in February 2011 when the collection system launched.
All of the tweets measure up to be 133.2 terabytes. That number includes two compressed copies of the 2006-2010 archives. "We were excited to be involved with acquiring the Twitter archives because it's a unique record of our time. It's also a unique way of communication," said Bill Lefurgy, digital initiatives program manager at the library.