Flowise
Open-source low-code platform to build LLM apps and AI agents visually with drag-and-drop chatflows and agentflows
Gallery
About Flowise
Flowise is an open-source, low-code platform for building LLM apps and AI agents visually, using drag-and-drop chatflows and agentflows instead of writing everything by hand. It suits individual developers and small teams prototyping ideas, businesses deploying real applications, and even non-technical users assembling chatbots and assistants. The visual builder lowers the barrier enough that you can have a working app running quickly.
The platform splits into a few clear building blocks. Chatflows cover single-agent systems with tool calling and RAG, agentflows handle multi-agent orchestration, and human-in-the-loop steps let people review or guide what agents do. It ships with a large set of integrations spanning LLMs, embedding models, and vector stores, plus APIs, SDKs, and an embeddable widget, so you avoid getting locked into one vendor and can drop your app into existing products. Execution tracing and observability help once things are in production.
It is licensed under Apache 2.0, so self-hosting is free, with a managed cloud option on a freemium model for teams that prefer not to run their own infrastructure. The active community and steady release pace make it a dependable choice for standing up grounded chatbots, agentic workflows, or private RAG search.
Key Features
- Drag-and-drop canvas for Chatflows and multi-agent Agentflows
- Connects 100-plus LLMs, embeddings, and vector stores across many providers
- Built-in RAG over TXT, PDF, DOC, CSV, and SQL sources
- Human-in-the-loop steps for review and approval inside a flow
- Deploy any flow as a REST API or an embeddable chat widget
- Self-host free via npm or Docker, or use managed Flowise Cloud
Pros & Cons
What we like
- Genuinely open source under Apache 2.0, so self-hosting is free
- Visual builder lets non-experts ship working LLM apps fast
- Broad model and vector-store support avoids lock-in to one vendor
- Large active community with 50k-plus GitHub stars and frequent releases
Room for improvement
- Self-hosting still needs your own model API keys and ongoing maintenance
- Cloud free tier is capped at 2 flows and 100 predictions a month
- Overage pricing past plan limits is not published, so costs are hard to predict
- Complex agent flows can get tangled and tricky to debug on the canvas
Frequently Asked Questions
What is Flowise?
Is Flowise free or open source?
Who is Flowise best for?
How does Flowise compare to LangChain?
Best For
Featured in
Alternatives to Flowise
View all
Dify
Open-source platform for building LLM apps and AI agents with visual workflows, RAG, and 100+ model providers
Gumloop
No-code canvas for building AI agents and automated workflows by connecting drag-and-drop nodes
LangChain
Open-source framework plus LangGraph and LangSmith for building, orchestrating, and observing LLM agents
CrewAI
Open-source Python framework for orchestrating role-playing multi-agent AI teams, with an enterprise platform
Reviews (16)
Good, with a few caveats
Started using Flowise casually, now it is pinned in my dock. Where it really wins is self-host free via npm or docker, or use managed flowise cloud. It fits well for standing up rag search over pdfs, spreadsheets, and knowledge bases. It would be a five if not for complex agent flows can get tangled and tricky to debug on the canvas.
Onboarded the team in a day
Picked Flowise for the price, stayed for the quality. Setup was painless and I was productive the same day. It fits well for self-hosting a private llm app stack to keep data in your own infrastructure. No regrets so far.
Good, with a few caveats
Flowise has quietly become part of my daily flow. The interface stays out of my way, which I appreciate. It has shaved real time off my week. Mostly using it for standing up rag search over pdfs, spreadsheets, and knowledge bases. It would be a five if not for self-hosting still needs your own model api keys and ongoing maintenance.
Genuinely impressed
Started using Flowise casually, now it is pinned in my dock. The defaults are sensible, so I was not fighting settings on day one. I expected to churn off it in a week and I am still here. Would sign up again without thinking twice.
Recommended without reservation
Started using Flowise casually, now it is pinned in my dock. Real selling point for me was genuinely open source under apache 2.0, so self-hosting is free. The core workflow is smooth once you are set up. It fits well for self-hosting a private llm app stack to keep data in your own infrastructure. Recommending it to people in a similar spot.
Quietly excellent
Hadn't planned on switching, but Flowise was hard to ignore. The connects 100-plus llms, embeddings, and vector stores across many providers is more useful than I expected. Mostly using it for prototyping agentic workflows that call tools and apis. Easy yes for anyone weighing the same trade offs.
Decent with some rough edges
Three months of Flowise later, here is what holds up. What stands out is how it handles drag-and-drop canvas for chatflows and multi-agent agentflows. Performance has been steady even when I lean on it hard. My only gripe is self-hosting still needs your own model api keys and ongoing maintenance. No regrets so far.
The kind of tool you forget you are paying for
Onboarded the whole team to Flowise in an afternoon. I expected to churn off it in a week and I am still here. It has been a fit for building ai chatbots and virtual assistants grounded in company docs.
Onboarded the team in a day
Almost a year on Flowise now, no plans to leave. Their take on drag-and-drop canvas for chatflows and multi-agent agentflows is genuinely good. Mostly using it for building ai chatbots and virtual assistants grounded in company docs. Recommending it to people in a similar spot.
It just works
Almost a year on Flowise now, no plans to leave. Genuine strength is that it gets out of the way and lets me work. I expected to churn off it in a week and I am still here. It fits well for standing up rag search over pdfs, spreadsheets, and knowledge bases. Glad I made the switch.
Good, with a few caveats
Onboarded the whole team to Flowise in an afternoon. What stands out is how it handles genuinely open source under apache 2.0, so self-hosting is free. My only gripe is overage pricing past plan limits is not published, so costs are hard to predict. Would sign up again without thinking twice.
Solid daily driver
Picked Flowise for the price, stayed for the quality. What stands out is how it handles deploy any flow as a rest api or an embeddable chat widget. The core workflow is smooth once you are set up. Found it works best for prototyping agentic workflows that call tools and apis.
Best decision this quarter
Tried Flowise on a side project first, then rolled it out everywhere. Real selling point for me was self-host free via npm or docker, or use managed flowise cloud. Would sign up again without thinking twice.
The kind of tool you forget you are paying for
Started using Flowise casually, now it is pinned in my dock. Their take on large active community with 50k-plus github stars and frequent releases is genuinely good. Mostly using it for self-hosting a private llm app stack to keep data in your own infrastructure. No regrets so far.
Genuinely impressed
Three months of Flowise later, here is what holds up. Where it really wins is broad model and vector-store support avoids lock-in to one vendor. The output quality holds up better than I expected. It has been a fit for building ai chatbots and virtual assistants grounded in company docs.
Recommended without reservation
Flowise has quietly become part of my daily flow. Their take on visual builder lets non-experts ship working llm apps fast is genuinely good. Nothing flashy, it simply pulls its weight every day.
Related Tools
LangChain
Open-source framework plus LangGraph and LangSmith for building, orchestrating, and observing LLM agents
Lindy
No-code AI agents that handle email, meetings, and recurring workflows
Wolli
Open framework for AI agents that grow around a purpose and extend themselves
n8n
Fair-code workflow automation with 400+ integrations