It was only a few days ago that we wrote about Boulder-based Gnip pulling in social data from China’s Sina Weibo. While those 300 million users do account for a huge amount of information, the growth of Tumblr makes Sina Weibo look like child’s play. Today, Gnip is announcing an exclusive partnership with Tumblr to pull in social data from Tumblr’s 50 million plus blogs.

For companies that want massive amounts of data that can be parsed into usable information, Tumblr’s astonishing 300% growth in 2011 is hard to ignore. Tumblr blogs rack up more than 15 billion page views each month, and the site has made itself a key source of social conversation, to the point that searches for Tumblr are expected to surpass searches for “blog” by October of this year.

What’s probably more impressive is that Tumblr’s Katherine Barna says that the deal is the first of its kind for Tumblr:

“This is the first time we have made this data available in a convenient, real-time format.”

According to Gnip COO Chris Moody, the goal is to be “the single source of record for all public conversations online.” When you consider Gnip’s data gets pulled from WordPress.com, Disqus, Facebook, YouTube, Twitter and many more, it’s easy to understand how the company is approaching what it wants to accomplish.

For Gnip, the different types of content on Tumblr (photos, text, quotes, links, etc) present an opportunity rather than a challenge. The company sees Tumblr as the place where younger trend-setters are discussing and sharing ideas, and it’s a chance to see what will be mainstream in the future.

Even before Tumblr, Gnip was delivering analysis on over 90 billion social media activities each month. While we don’t have a measurement on exactly how many activities the addition of Tumblr will create, it’s nothing short of staggering, and a highly-valuable tool for brands who are measuring sentiment.