This article presents a brief overview of the end-to-end conversation modeling track of the 6th dialog system technology challenges (DSTC6). The task was to developa fully data-driven dialog system using a customer service conversations from Twitter business feeds, and invited participants who work on this challenge track. In this overview, we describe the task design and data sets, and review the submitted systems and applied techniques for conversation modeling. We received 19 system outputs from six teams, and evaluated them based on several objective measures and a human-rating based subjective measure. Finally, we discuss technical achievements and remaining problems related to this challenge.