google-deepmind/concordia
Concordia is a library for running multi-agent simulations where LLM-powered entities interact in environments controlled by a Game Master.
Not currently ranked — collecting fresh signals.
star history
Concordia is a Python library for constructing and running generative agent-based models that simulate interactions among entities in physical, social, or digital environments. It uses a tabletop role-playing game pattern where a Game Master entity manages the simulation environment and translates natural-language actions into outcomes. The library requires access to standard LLM APIs and supports applications including social science research, AI safety evaluation, synthetic data generation, and performance assessment.
Frequently asked
- What is google-deepmind/concordia?
- Concordia is a library for running multi-agent simulations where LLM-powered entities interact in environments controlled by a Game Master.
- Is concordia open source?
- Yes — google-deepmind/concordia is open source, released under the Apache-2.0 license.
- What language is concordia written in?
- google-deepmind/concordia is primarily written in Python.
- How popular is concordia?
- google-deepmind/concordia has 1.6k stars on GitHub.
- Where can I find concordia?
- google-deepmind/concordia is on GitHub at https://github.com/google-deepmind/concordia.