LELú is a French dialog corpus that contains a rich collection of human-human, spontaneous written conversations, extracted from Reddit’s public dataset available through Google BigQuery. Our corpus is composed of 556,621 conversations with 1,583,083 utterances in total. The code to generate this dataset can be found in our [GitHub Repository][1]. The archive `spf.tar.gz` contains Reddit discussions in an XML file with the following format: The tag attributes can be described as follows: - `link_id`: ID of the parent Reddit post. - `subreddit_id`: ID of the subreddit. - `uid`: ID of the comment author. - `comment_id`: ID of the Reddit comment. - `parent_id`: ID of the parent Reddit comment. We have split up the conversation trees into short sequential conversations using a heuristic described in our paper, [LELú: A French Dialog Corpus from Reddit][2], however the full conversation trees can be reconstructed using the `comment_id` and `parent_id` attributes of the `