Open-source Conversational AI Stack

At DeepPavlov, our work contributions are spread across 3 parts of the technology stack, including Library, Dream, and Agent. All of these projects are focused on enhancing and strengthening our Conversational AI stack, with each contributing corresponding components.

DeepPavlov Library is a foundation for our framework. DeepPavlov is an Open-source framework for building chatbots and virtual assistants. It comes with a set of predefined components for solving Natural Language Processing (NLP) related problems and a framework to build a modular pipeline that lets developers and NLP researchers create production-ready conversational skills. DeepPalvov Library is based on TensorFlow, PyTorch, and Keras. It contains basic NLP components like NER, Entity Linking, KBQA, Go-Bot, and many others.

DeepPavlov Agent is our multiskill Conversational AI orchestrator that coordinates the entire Conversational AI pipeline of the AI Assistants. It incorporates annotators, skills, Skill & Response Selectors to provide a coherent experience to its users. It was used by our socialbots like one we've built for Amazon Alexa Prize 3 and our DREAM socialbot, by Deepy - our Open-source Multiskill AI Assistant, and is currently powering our socialbot built for Amazon Alexa Prize 4.

DP Library and DP Agent make it easy for beginners and experts to create advanced dialog systems. Additionally, its declarative approach to defining sequences of model execution in config files allows users to track dependencies and provide paths to download the missing trained ML models.

DeepPavlov Dream is a set of our default goal-oriented and chit-chat skills, as well as a number of demo AI Assistants built using components from Library and managed by DeepPavlov Agent.

lightbulb_outline View ideas list


  • python
  • docker
  • python deep learning frameworks
  • tensorflow
  • pytorch


  • Other
  • nlp
  • conversational
  • agent
comment IRC Channel
email Mailing list
mail_outline Contact email

DeepPavlov 2021 Projects

  • Anshuman Singh
    Refactor Multi-Task BERT
    Multi-task learning shares information between related tasks, reducing the number of parameters required. State of the art results across natural...
  • Anastasiia Sedova
    Relation Extraction
    The main goal of the project is to provide the user of DeepPavlov Framework with an out-of-box solution for relation extraction. I would consider...
  • Niklas Muennighoff
    TripPy + DeepPavlov
    The TripPy architecture brings transformer models to Goal-oriented Chatbots. While setting new SoTA results on MultiWOZ among others, TripPy also...