CrewAI is an open-source Python framework for building and orchestrating multi-agent AI systems where each agent is assigned a role, goal, and backstory, and tasks are executed collaboratively within a 'crew.' It supports sequential, hierarchical, and consensus-based process flows. The managed platform (CrewAI AMP) adds a visual no-code editor, AI copilot, centralized management, and serverless execution. As of 2026, the framework has 47,800+ GitHub stars and reportedly powers around 450 million agentic workflow executions per month across Fortune 500 customers. Key features: - Role-based agent definition with goal, backstory, and task assignment - Sequential, hierarchical, and consensual workflow process types - CrewAI AMP: visual no-code crew builder with AI design copilot - Serverless agent execution with role-based access control - Workflow tracing, agent training, and task guardrails - Open-source framework (MIT license) with no usage limits or artificial caps
Open-source framework: free. CrewAI AMP platform: Free (50 executions/month); Professional $25/month (100 executions/month, $0.50/overage); Enterprise custom (~$60,000–$120,000/year estimated).