AI workflows and agents # Build AI agents with Flowise without writing code Deploy Flowise on Deplomo, a visual platform based on LangChain for designing AI workflows, RAG systems and agents. [Get started](https://panel.deplomo.com/signup)[Back to marketplace](/marketplace) ## Flowise features on Deplomo Visual workflow editor, persistent storage and connections to supported AI providers. ### Build agents visually without code Design multi-step AI workflows, processing chains and agents with a drag-and-drop interface. ### Deploy Flowise on the cloud platform Deploy Flowise from a template and configure CPU, RAM and storage for your workload. ### Connect to vector databases and knowledge bases Connect vector databases to build retrieval systems and retrieval-augmented generation (RAG) workflows. ### Connect to models and external APIs Connect to commercial and open-source models such as OpenAI, Claude and Ollama, along with webhooks and custom tools. ## Flowise FAQs Answers to common questions about building agents, connecting knowledge bases and deploying Flowise: What is Flowise, and how is it used to build AI applications? What are the benefits of building agents with a visual interface? How can I add Flowise chatbots and agents to a website or app? Does Flowise support Persian and custom models? Why host Flowise on Deplomo instead of running it locally? ## Flowise guide: build AI agents with a visual workflow editor Building AI systems can involve coding with frameworks such as LangChain, managing memory pipelines and debugging prompts. Flowise provides a visual, modular interface for creating AI workflows, chatbots connected to documents and automated agents. Availability and service security depend on your deployment and configuration. ### What is Flowise, and how does its visual interface work with LangChain? Flowise provides a visual interface for LangChain workflows. Instead of writing code to define chains, tools, memory and language models, connect blocks on a canvas. To build semantic search and retrieval-augmented generation (RAG) workflows, add a vector database to the canvas. For example, connect the [qdrant](/marketplace/apps/qdrant) in Flowise to index documents and answer questions based on their content. To connect business processes and webhooks to AI workflows, use an automation tool such as [Install n8n](/marketplace/apps/n8n) alongside Flowise. ### Flowise running locally compared with cloud deployment Running Flowise on a local system or unmanaged server can make persistence, process monitoring and access to external language model providers your responsibility. Flowise runs in an isolated container and can use persistent storage. Connections to providers such as OpenAI and Anthropic require the relevant API credentials and remain subject to each provider’s access rules and availability. ### Compare Flowise on Deplomo with manual coding and traditional server setup The table compares different ways to use Flowise:
| Feature | Flowise on Deplomo | Manual coding with LangChain | Running locally on a personal computer |
|---|---|---|---|
| Design and prototyping | Build an agent visually without writing code | Write code and debug libraries | May work for initial tests but require additional work for production use |
| Service availability and reliability | Runs in a cloud container with persistent storage | Requires you to maintain and monitor the server | Stops when the local machine is shut down |
| Embed the widget and use its output | Generate a JavaScript widget and API for your website | Requires separate backend and frontend development for chat | May require additional setup to share publicly |
| External model provider access | Use your API credentials and verify provider access | May require proxy configuration in code | Depends on local network and provider access |