Abstract
We present BlenderBot 3, a 175B parameter dialogue model capable ofopen-domain conversation with access to the internet and a long-term memory,and having been trained on a large number of user defined tasks. We releaseboth the model weights and code, and have also deployed the model on a publicweb page to interact with organic users. This technical report describes howthe model was built (architecture, model and training scheme), and details ofits deployment, including safety mechanisms. Human evaluations show itssuperiority to existing open-domain dialogue agents, including its predecessors(Roller et al., 2021; Komeili et al., 2022). Finally, we detail our plan forcontinual learning using the data collected from deployment, which will also bepublicly released. The goal of this research program is thus to enable thecommunity to study ever-improving responsible agents that learn throughinteraction.