TL;DR
Langflow 1.5 brings a simplified, stable foundation for building AI workflows. With clean core components, improved execution, and a smoother UI, Langflow keeps evolving for real-world use. Langflow Desktop is now available for Windows, and the new Docling Bundle takes document processing to a new level.
A Clean, Stable Core
Langflow 1.5 introduces a more focused set of core components, simplifying how you build flows. Redundant and overlapping components have been removed, making the experience clearer and more predictable. These curated components are designed to serve as the foundation for modular, reliable flows, and lay important groundwork for potential features such as AI-assisted development inside Langflow.
Along with the cleanup, this release brings improved components that make it easier to connect and process information. Components like News Search, RSS Feed, and Web Search add new ways to bring external data into flows and quickly validate ideas. For processing tasks, components such as Data and DataFrame operations, the LLM Router, and Smart Functions enable more flexible logic and powerful data handling.
MCP Usability Upgrades
We’ve made it easier to integrate Langflow with the tools you already use. MCP integration is now more straightforward, with one-click setup for popular clients like Claude, Cursor, and Windsurf. Tool configuration has also been improved, with clearer action settings and editing options. Inside the canvas, dynamic output handling and dropdowns in components make iteration faster and building more intuitive.
Smarter Document Processing with Docling
Langflow 1.5 introduces the Docling Bundle, making it simple to process diverse document formats and extract structured information for your AI workflows.
Docling supports parsing of PDFs, DOCX, PPTX, XLSX, HTML, and more. It also provides advanced PDF understanding, including reading order, layout, tables, code, formulas, and image classification.
It’s all designed to make RAG pipelines work better! Behind the scenes, Docling uses a structured, JSON-serialized document format. This allows downstream components, like the Docling Chunker, to apply structure-aware chunking based on the document hierarchy, instead of relying on raw text alone.
Langflow Desktop Now on Windows
Langflow Desktop is now available for Windows! You can build, test, and run flows entirely on your Windows machine. We're shipping support for x86 now and support for Arm is coming soon.
Under the Hood
A big part of this release is less visible, but just as important. Langflow 1.5 includes significant work under the hood to make installing, building and running flows more stable.
Flow execution has been improved to run more smoothly, whether you're testing locally or triggering flows through the API. At the same time, we’ve addressed dozens of bugs and common pain points reported by our community!
Speaking of which…
Langflow is open-source and evolving fast!
- Join our Discord community to share feedback, get support, or connect with other builders.
- Check out the GitHub repo to explore, contribute, or star the project 🌟
- Subscribe to our Youtube channel for tips and trick on using Langflow and building apps.
We can't wait to see what you build with Langflow. Download 1.5 and let us know what you think!
