App hosting
Launch Flowise on NABHCLOUD in one click. Build chatbots, document search tools and agents, secured with HTTPS and billed with GST invoicing included.
Flowise is a visual canvas for building LLM-based tools — support chatbots, document search bots and multi-step agents — assembled from LangChain building blocks without any coding.
Deploy it straight from our app marketplace, then check our hosting plans for current configurations and pricing.
Flowise needs at least our Starter tier. Every plan bundles HTTPS, support for a custom domain and a GST-compliant invoice, with servers located in Noida.
Installing Flowise brings it online over HTTPS right away. You'll still need to connect a chat-model credential — an OpenAI-compatible provider hosted in India keeps both your data and your invoices within the country.
Flowise provides a drag-and-drop workspace for LLM pipelines. Connect a model for chat, a loader for your documents, a splitter that breaks text into chunks, a model that generates embeddings and a store for vectors, linking them visually to produce a chatflow that exposes a callable endpoint plus a widget you can embed. Internally it runs on Node.js, built from LangChain and LlamaIndex pieces, distributed under the Apache 2.0 licence, and it listens on port 3000.
Getting a first build running tends to be fast. Complications tend to surface in the second week instead. Flowise stores its database (SQLite), any uploaded files and the encryption key protecting saved credentials inside a folder called .flowise. If that key gets lost, every stored credential will need to be re-entered, so that folder deserves a place in your backup routine. A separate pitfall is the default vector store, which lives only in memory and clears itself on every restart — switch to a persistent store before importing documents you actually care about.
By default, any chatflow becomes reachable by anyone who has its link. Require an API key before your own code is allowed to call a flow, and cap which domains may load the embedded widget — that stops others from riding on your bot and running up your bill.
| Workload | Plan | Note |
|---|---|---|
| Testing prompts and running a small chatbot | Starter | Most of the processing happens on the model's API, not on your server. |
| Running several flows with document uploads and multiple users | Pro | Parsing and splitting documents locally consumes meaningful memory. |
Our Starter configuration offers 1 vCPU and 1 GB of RAM, with 20 GB of disk space, enough room for up to 2 apps. It's sized so Flowise can sit alongside your own application on one server, billed monthly before 18% GST.
Ready to proceed? Install Flowise from the app marketplace.
Whatever provider Flowise already supports. If you want an OpenAI-compatible endpoint hosted within India, point it at that provider's base URL using your own key.
Yes — Flowise generates either a script tag or an iframe snippet for each chatflow you build.
Everything lives in a folder named .flowise — it holds the SQLite database, your uploaded files, and the key that encrypts stored credentials. Back up that whole folder.
Yes, if the chatflow is public, or if its embedded widget has no domain restriction. Add an API key and set allowed domains per chatflow to block that access.
Pick an app from the marketplace, choose which application server will host it, and select Install. It receives its own HTTPS address immediately, its passwords are generated and shown to you once, and it's live within a couple of minutes.
Tell us what you plan to run and we will suggest the right setup.