Key Takeaways
- Apache 2.0 license β the self-hosted code is free and stays usable indefinitely, even archived
- Active development stopped July 29, 2026; the GitHub repository was archived (read-only) on August 10, 2026
- Flowise Cloud, the paid hosted version, shuts down completely on November 30, 2026
- Two build modes: Chatflow (single-agent chatbots with RAG and tool calling) and Agentflow (multi-agent orchestration)
- Self-hosting via Docker or npm still works today, but with no security patches going forward
- For any new project with a multi-year horizon, evaluate a maintained alternative or a community fork instead of starting fresh on Flowise
π In One Sentence
Flowise is an open-source, Apache 2.0, drag-and-drop builder for LangChain/LangGraph agents and RAG pipelines, but the company behind it is winding down and Flowise Cloud shuts down November 30, 2026.
π¬ In Plain Terms
Instead of writing LangChain code by hand, Flowise gave you visual boxes and arrows to connect an LLM, a document loader, and a vector store β but no one is fixing bugs or reviewing pull requests anymore.
β οΈWarning: Do not sign up for Flowise Cloud. The hosted product is scheduled to shut down on November 30, 2026, and support already ended on August 31, 2026.
Is Flowise Still Maintained?
No. FlowiseAI announced on July 29, 2026 that it is winding down the company, citing a shift toward AI coding agents (such as Claude Code) that handle complex tasks more flexibly than a node-based visual workflow tool. The shutdown follows a fixed schedule with four dated milestones.
The GitHub repository remains publicly visible under the Apache 2.0 license "indefinitely," according to FlowiseAI, and the team explicitly encourages users to fork the project and maintain their own copy. No successor product or acquiring company has been named.
Development freeze
- Date:
- 2026-07-29
- What it means:
- No new features; pull requests no longer reviewed or merged
Repository archived
- Date:
- 2026-08-10
- What it means:
- GitHub repo is read-only; npm packages and Docker images marked deprecated
Support ends
- Date:
- 2026-08-31
- What it means:
- Core team leaves Discord and GitHub issue threads; community moderation only
Cloud shuts down
- Date:
- 2026-11-30
- What it means:
- Flowise Cloud, the paid hosted product, stops running entirely
πNote: This changes the calculus for every recommendation below: self-hosting the archived Apache 2.0 code is viable today, but Flowise Cloud is not β regardless of what pricing pages still show.
What Is Flowise?
Flowise is an open-source platform (Apache 2.0 license, github.com/FlowiseAI/Flowise) for building AI agents and LLM applications through a visual, node-based interface instead of writing code directly. Its tagline was "Build AI Agents, Visually," and the repository passed 55,000 GitHub stars before development stopped.
- Chatflow: single-agent chatbots with retrieval-augmented generation (RAG) and tool calling, built by wiring nodes together
- Agentflow: multi-agent systems with workflow orchestration between several cooperating agents
- HITL (Human In the Loop): pauses a flow for manual review or approval before it continues
- Support for 100+ LLM providers, embedding models, and vector databases through pre-built nodes
- Deployment as an API endpoint, an embeddable chat widget, or through an SDK
- Execution traces with observability hooks for Prometheus and OpenTelemetry
How Does Flowise Differ From LangChain and LlamaIndex?
Flowise is a visual layer that generates the same underlying LangChain and LangGraph constructs a developer would otherwise write in Python or JavaScript. LangChain and LlamaIndex are code-first frameworks: you import a library, write chain logic, and control every step programmatically. Flowise trades some of that flexibility for faster visual iteration and a shallower learning curve.
Does Flowise Support RAG and AI Agents?
Yes, both are first-class in Flowise, and this was the platform's core use case before development stopped.
- RAG: document loader nodes (PDF, web scraper, CSV, and more) feed a text splitter, an embedding model node, and a vector store node β Chatflow assembles the retrieval pipeline visually
- Agents: Agentflow adds tool-calling nodes, conditional branching, and multi-agent handoff so several specialized agents can pass a task between each other
- Memory: conversation memory nodes persist chat history across turns, scoped per session or per user
- Tool calling: agents can call external APIs, run code, or query a database through pre-built or custom tool nodes
How Do You Self-Host Flowise Now?
Self-hosting from the archived repository is the only path that makes sense today β do not create a Flowise Cloud account. These steps run Flowise locally or on your own server using Docker, the fastest supported method before archival.
- 1Install Docker and Docker Compose if you do not already have them, on Linux, macOS, or Windows with WSL2.
- 2Clone the archived repository: git clone https://github.com/FlowiseAI/Flowise.git, then cd Flowise/docker.
- 3Copy the example environment file (cp .env.example .env) and set FLOWISE_USERNAME, FLOWISE_PASSWORD, and any LLM provider API keys you plan to use.
- 4Start the stack with docker compose up -d. Flowise listens on port 3000 by default; change PORT in .env if that conflicts with another service.
- 5Open http://localhost:3000, sign in with the credentials from step 3, and build a Chatflow or Agentflow by dragging nodes from the sidebar onto the canvas.
- 6Back up your Flowise data directory regularly. With no upstream maintainer, a corrupted local database has no vendor support to fall back on.
Does self-hosted Flowise still work after the archival?
Yes. The archived code runs exactly as it did before July 29, 2026 β archival stops new commits and pull request review, it does not disable the software you already have running or the copy you clone today.
Will Flowise self-hosting receive security patches?
No. The GitHub repository is read-only and the core team has left support channels, so no one is triaging or patching newly discovered vulnerabilities. Treat a self-hosted instance the same as any unmaintained open-source dependency: isolate it, keep it off the public internet where possible, and monitor its dependencies.
Who Should Use Flowise?
The shutdown changes who Flowise makes sense for. It is no longer a safe default for a new production project, but it still has a narrow, legitimate use case.
Flowise vs. Alternatives
For anyone starting a new project today, an actively maintained alternative removes the biggest risk Flowise now carries: no one fixing bugs.
| Tool | Interface | License | Best For | Maintenance |
|---|---|---|---|---|
| Flowise | Visual drag-and-drop | Apache 2.0 | Existing self-hosted setups | Archived / none |
| LangChain | Python / JS code | MIT | Full custom control | Active |
| LlamaIndex | Python / TS code | MIT | Data-heavy RAG pipelines | Active |
| n8n | Visual drag-and-drop | Fair-code / Sustainable Use | General workflow automation | Active |
Common Mistakes When Evaluating Flowise Now
These mistakes come from treating pre-shutdown Flowise information β including the vendor's own marketing pages β as still current.
Frequently Asked Questions
Is Flowise still maintained?
No. FlowiseAI stopped active development on July 29, 2026, archived the GitHub repository (read-only) on August 10, 2026, and the core team left Discord and GitHub support channels on August 31, 2026.
Can I still self-host Flowise?
Yes. The archived repository is public under Apache 2.0 and runs the same as before. You can clone it, run it with Docker, and use every existing node β you just will not get bug fixes, security patches, or new integrations going forward.
What happens to Flowise Cloud?
Flowise Cloud, the paid hosted version, shuts down completely on November 30, 2026. Do not sign up for it or rely on it for a new project.
What license is Flowise released under?
Apache License 2.0. This lets anyone use, modify, and redistribute the code, including forking it into an independently maintained project, which is exactly what the FlowiseAI team recommended to its users.
How is Flowise different from LangChain?
Flowise is a visual, node-based interface that generates the same LangChain and LangGraph building blocks a developer would otherwise write in code. LangChain itself is a code-first framework with no visual canvas β it stays actively maintained independently of Flowise's shutdown.
Does Flowise support retrieval-augmented generation (RAG)?
Yes. Chatflow mode connects document loader, text splitter, embedding model, and vector store nodes to build a RAG pipeline visually, without writing retrieval code by hand.
Does Flowise support multi-agent systems?
Yes, through Agentflow, which adds tool-calling nodes, conditional branching, and hand-off between multiple cooperating agents in a single canvas.
Why is FlowiseAI shutting down the company?
The team cited a broader shift toward AI coding agents like Claude Code, which they say handle complex, non-linear tasks more flexibly than a rigid node-based visual workflow tool once a project grows past simple pipelines.
What should I use instead of Flowise for a new project?
For code-first control, LangChain or LlamaIndex remain actively maintained. For a visual, no-code builder with ongoing development, evaluate n8n or a community fork of Flowise once one proves stable β do not default to unforked, archived Flowise for anything long-term.
Is it safe to self-host an archived, unmaintained project?
It carries the same risk as any unmaintained open-source dependency: no one is patching newly discovered vulnerabilities. Keep it off the public internet where possible, isolate it from sensitive systems, and monitor it the way you would any dependency with no active maintainer.