CAMEL
CAMEL-AI.orgCAMEL is a Python framework for building and studying multi-agent systems, from pairs of role-playing agents to larger agent societies and workforces. It provides chat agents with stateful memory, toolkits, retrieval, storage and model integrations, and is used for synthetic data generation, task automation and world simulation. CAMEL-AI.org, a community research group, builds it; its 2023 paper introduced the role-playing approach.
Openness
5 high confidence- license
- Apache-2.0 (OSI)
- source
- public
- core features withheld
- no — the tree at 0366b64 holds 2,243 paths under one Apache-2.0 LICENSE with no ee/, enterprise/, commercial/ or proprietary/ path; the README presents the project as a community research collective and lists Eigent, the team's commercial multi-agent product, among separate products built with CAMEL
- commercial
- Eigent
CAMEL's source is Apache-2.0 licensed and builds from its public repository with no enterprise directory. The team also sells Eigent, a separate multi-agent product built on CAMEL, which does not take anything out of the open framework.
- https://pypi.org/pypi/camel-ai/json recorded 2026-11-12
PyPI metadata for camel-ai 0.2.90 declares license_expression Apache-2.0 and names https://github.com/camel-ai/camel as its repository; pre-releases of 0.2.91 follow through 2026-09-03.
- https://raw.githubusercontent.com/camel-ai/camel/master/LICENSE recorded 2026-11-12
The LICENSE is the Apache License, "Version 2.0, January 2004", with the appendix line "Copyright 2023 @ CAMEL-AI.org".
- https://raw.githubusercontent.com/camel-ai/camel/master/README.md recorded 2026-11-12
The README says "The source code is licensed under Apache 2.0", calls the team "a community-driven research collective", and lists Eigent under products "Built with CAMEL".
Adoption
2 medium confidenceAdoption is read from PyPI downloads of camel-ai over the trailing month. Automated installs add to the count, so it measures install volume rather than researchers or developers.
- https://pypistats.org/api/packages/camel-ai/recent recorded 2026-11-12
last_month 42,975 downloads for camel-ai (last_week 11,980, last_day 1,591).
Capability
4 medium confidenceCAMEL covers multi-agent societies, workforces, memory, retrieval and a wide set of toolkits, with a research focus on generating data and simulating agent populations. It sits a step below LangGraph, whose runs checkpoint their state and resume where they stopped.
- https://raw.githubusercontent.com/camel-ai/camel/master/README.md recorded 2026-11-12
The README lists Agent Societies, Memory, Storage, Retrievers and toolkits as modules, the Workforce multi-agent system, RAG and graph RAG cookbooks, and data generation, task automation and world simulation as uses; it publishes no benchmark figure.
Verified 2026-11-12