# Langflow A low-code builder for developers creating AI agents, RAG applications, and MCP servers. - Findings24: - Site: - Preços: Freemium - Categorias: Construtor de Apps IA, Criador de agentes de IA, Ferramentas MCP ## O que é Langflow? Langflow is a low-code AI builder for creating and deploying agentic and retrieval-augmented generation (RAG) applications, as well as Model Context Protocol (MCP) servers. It combines a visual flow editor with Python customization, letting developers connect models, data sources, and tools into working applications. Its core workflow moves from assembling reusable components to running flows and exposing them through an API. Developers can choose models and vector databases for their application, iterate on the flow, and deploy through either the open-source software or a cloud account. ## O que Langflow pode fazer? ### Visual flow building Assemble workflows with drag-and-drop components and visual state flows. Hundreds of pre-built flows and components provide starting points for building and iterating on applications. ### Python customization Customize components and application behavior in Python when a workflow needs logic beyond the visual configuration. ### Model and data integrations Swap and compare models and connect vector databases and external tools. Listed integrations include Anthropic, Ollama, Mistral, Qdrant, Weaviate, Google Drive, and Slack. ### Agent tool access Run a single agent or multiple agents, with flow components available as tools. Agent configuration includes instructions, a model provider, and tool inputs. ### Structured output generation Use a language model to extract structured data from unstructured text with a defined output schema, supporting workflows that need consistently formatted results. ### API deployment Expose a flow as an API to connect it with other applications. Langflow supports self-deployment and cloud deployment for moving workflows into production. ## Casos de uso ### Call transcription and support analysis Convert audio files into transcripts with AssemblyAI, then classify calls by topic, sentiment, resolution status, and urgency. The workflow preserves both the transcript and its classification for subsequent analysis. ### Document extraction Turn unstructured PDFs, emails, and web pages into machine-readable data. Combine document processing with schema-based output generation to prepare extracted information for downstream workflows. ### Document chunk classification Ingest files, split their contents into chunks, and classify those chunks in a document-processing workflow for documents and knowledge bases. ### CSV query assistance Build an assistant that processes uploaded files for automated document analysis and data queries, using the CSV Query Assistant template as a starting point. ## Perguntas frequentes ### Do I need to write Python to start building? Langflow offers a visual editor and pre-built components for assembling flows. Python customization is available when you need to modify components or add application-specific behavior. ### How do I configure a model in a flow? Language model components let you select a provider and model, enter an API key, and supply input and system messages. The published examples show OpenAI model configuration. ### Can I inspect a flow's chat output while building? Yes. A Chat Output component displays messages in the Playground, allowing you to view the output of a connected flow. ### Is there a free way to get started? The website offers free getting started access and a free cloud account, alongside the open-source option for self-deployment. ## Informações rápidas - **Product category**: Low-code AI application builder - **Primary users**: Developers building AI workflows - **Customization language**: Python - **Application types**: AI agents, RAG applications, and MCP servers - **Website**: https://langflow.org