Skip to main content
Flowise is an open-source drag-and-drop UI tool that lets you build custom LLM apps in minutes. This guide shows you how to integrate SambaCloud’s LLMs with Flowise to power your workflow agents with fast inference.
This guide was tested with Flowise v2.x. If you’re using a different version, some UI paths or features may vary.

​
Prerequisites

​
Setup

1

Install Flowise

On newer Node versions (v25+), npm install -g flowise may fail with a better-sqlite3 native build error. If you hit this, use the Docker container instead:
Then skip the npx flowise start step below and open http://localhost:3000 directly.
2

Start Flowise

3

Open the dashboard

Open your browser to http://localhost:3000. You should see the Flowise dashboard with the Chatflows panel.

​
Usage

​
Create a chat flow

In the Chatflows panel, click Add New in the upper-right corner.
Flowise Chatflows panel

​
Add a SambaNova node

  1. In the left sidebar, expand the Chat Models category.
  2. Drag the ChatSambaNova node onto the canvas.
Flowise node selection

​
Configure SambaNova credentials

  1. Click the ChatSambaNova node on the canvas.
  2. In the Credential dropdown, select Create New.
  3. Enter your SambaCloud API key and click Save.
  4. Select a model from the Model Name dropdown.
Flowise credentials panel

​
Build your workflow

Connect additional nodes to create your workflow. Common configurations include:
  • Basic chat: Connect the ChatSambaNova node directly to a chat output.
  • RAG pipeline: Add document loaders, vector stores, and retrieval chains.
  • Agents: Combine with tool nodes for agentic workflows.

​
Test the chat

  1. Click the Chat icon in the upper-right corner of the canvas.
  2. Send a test message to verify the connection.
Flowise chat panel

​
Capabilities

Once connected, you can use SambaCloud models in various Flowise workflows:
  • Conversational agents: Build chatbots with memory and context awareness.
  • Document Q&A: Create RAG pipelines that ground responses in your documents.
  • Tool-using agents: Connect external tools and APIs for agentic workflows.
  • Custom chains: Design multi-step LLM pipelines with branching logic.

​
Troubleshooting

You’re likely on Node v25+. Use the Docker container instead:
Then open http://localhost:3000 directly — skip the npx flowise start step.
Ensure you’re running Flowise v2.x or later. Restart Flowise after updating.
Verify your API key is correct. Regenerate it in the SambaCloud console if needed.
Check your network connection and ensure api.sambanova.ai is accessible.
Confirm you’ve selected a valid model from the dropdown and saved your credentials.

​
Additional resources