Flowise AI

Flowise is an open source platform for building AI agents and LLM-powered applications through a visual, drag-and-drop interface. It targets developers and teams who want to prototype and ship agentic systems without writing orchestration logic from scratch. The core idea is modular: you connect building blocks together visually, then deploy the result as a production-ready API or embedded chat widget.

There are two main flow types. Agentflow handles multi-agent systems where work is distributed across coordinated agents, each with defined roles and tools. Chatflow covers single-agent setups and chatbots, including tool calling and retrieval-augmented generation (RAG) from external data sources. Both support Human in the Loop (HITL), letting a human review or approve agent actions before they proceed.

Flowise connects to over 100 LLMs, embedding models, and vector databases. Finished flows are accessible via REST API, a TypeScript or Python SDK, or an embeddable chat widget. That makes it practical to drop AI functionality into an existing product without rebuilding infrastructure around it. If you've used something like Langflow or Dify, the general approach will feel familiar, though the feature sets differ.

Observability is built in. Full execution traces are available for every run, and the platform integrates with Prometheus and OpenTelemetry for teams that need production monitoring. Horizontal scaling is supported through message queues and workers, and deployment works in both cloud and on-premises environments.

The self-hosted version is fully open source. Flowise also offers a managed cloud with a free tier (2 flows, 100 predictions/month) and paid plans starting at $35/month for individuals who want unlimited flows. The Pro plan adds workspaces, role-based access control, and priority support for larger teams.

Flowise has recently joined Workday, though the open source project continues to be developed and maintained.