Flowise

What it does

Flowise is an open source generative AI development platform for building AI agents and LLM workflows. Builders assemble modular components on a visual canvas, connecting models, embeddings, vector databases, and data sources. Its three visual builders are Assistant, Chatflow, and Agentflow, and it accepts uploaded documents in formats including PDF, DOC, CSV, and Markdown for retrieval-augmented generation (RAG).

People typically use Flowise to prototype and deploy chatbots, document question-answering assistants, and multi-agent workflows without writing extensive code. The platform adds execution traces, human-in-the-loop review, and MCP client/server nodes, and can be consumed through its API, JavaScript and Python SDKs, CLI, or an embeddable chat widget.

Flowise suits developers and technical teams building agents and retrieval applications. It offers hosted cloud access and a self-hosted installation using Node.js or Docker. Self-hosting requires managing the runtime, security, backups and updates. Cloud plans meter predictions and storage, and model-provider charges and credentials need separate attention. Check the current installation guide for supported Node.js versions before deploying.

What it can do

• Visual drag-and-drop builder for chatbots, assistants, and multi-agent workflows
• Retrieval-augmented generation over uploaded files and 100+ data sources
• Tool calling and branching, looping, and routing logic
• Human-in-the-loop review of agent tasks
• Execution traces with Prometheus and OpenTelemetry support
• Embeddable chat widget plus REST API and TypeScript/Python SDKs

Works with

• 100+ LLMs, embeddings, and vector databases
• MCP client and server nodes
• Prometheus
• OpenTelemetry
• LangChain

Example workflows

Document Q&A chatbot

Illustrative example. Input: PDF files uploaded to a Chatflow. Steps: connect an LLM and vector database, index the documents, add a retriever, and test in the chat panel. Output: a chatbot that answers questions grounded in the uploaded files.

Embedded support widget

Illustrative example. Input: a configured Assistant with a knowledge base. Steps: enable the embedded chat widget, customize branding, and place the snippet on a website. Output: a branded chat widget on the site that answers visitor questions.

Before you choose

• Self-hosting requires managing servers, database backups, and updates; cloud is recommended for non-technical users.
• Cloud plans limit flows, predictions per month, storage, and users; check the pricing section for current caps.
• Enterprise deployment and SSO require contacting Flowise support at [email protected].

Listed under

• Coding
• App builders
• AI agents
• Chatbots
• Developers
• Developer tools
• LLM apps
• Multi-agent

Get started

Sign in to the hosted app, or install locally with npm install -g flowise.

1. Choose Flowise Cloud, or check the current Node.js and Docker requirements for a self-hosted installation.
2. Run npm install -g flowise, then npx flowise start, and open http://localhost:3000.
3. Connect an LLM and vector database in the visual builder before building a flow.

The open-source platform can be self-hosted; infrastructure and model-provider costs remain separate. Cloud Starter is $35/month and Pro $65/month, with prediction, storage and user limits. Confirm extra-seat charges and any first-month promotion at signup.

Source: flowiseai.com, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Flowise - Build AI Agents, Visually ↗flowiseai.com
• Get Started | FlowiseAI ↗docs.flowiseai.com
• Flowise - Build AI Agents, Visually ↗flowiseai.com
• Introduction | FlowiseAI ↗docs.flowiseai.com

Company website

flowiseai.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.