Framework for productionizing multi-agent workflows
Top 63.3% on sourcepulse
FastAgency is an open-source framework designed to accelerate the transition from prototype to production for multi-agent AI workflows, specifically targeting developers using the AG2 (formerly AutoGen) framework. It provides a unified interface to scale Jupyter notebook prototypes into production-ready applications, supporting various UIs and distributed system deployments.
How It Works
FastAgency offers a unified programming interface that allows developers to write workflows once and deploy them across different user interfaces (ConsoleUI, MesopUI) and network adapters (FastAPI, NATS.io via FastStream). This approach simplifies the development lifecycle by enabling code reuse and abstracting away deployment complexities, whether for local testing, web applications, or distributed, multi-datacenter systems. It also facilitates seamless integration of external APIs via OpenAPI specifications.
Quick Start & Requirements
cookiecutter https://github.com/ag2ai/cookiecutter-fastagency.git
.OPENAI_API_KEY
).pytest
.Highlighted Details
Tester
class for continuous integration of agent workflows.Maintenance & Community
Licensing & Compatibility
Limitations & Caveats
Currently, FastAgency only supports the AG2 (formerly AutoGen) runtime. The license is not explicitly stated, which may impact commercial adoption or integration with closed-source projects.
5 days ago
Inactive