Organizations everywhere are racing to build AI systems powered by their internal data. But turning documents into reliable context for AI remains one of the hardest challenges when moving applications from experimentation to production.
OpenRAG was created to solve this problem.
With the General Availability of OpenRAG 0.4 and the launch of IBM Elite Support for OpenRAG, developers and enterprises now have an open agentic RAG tool designed to transform unstructured data into structured context for RAG and agentic AI applications with enterprise support.
What’s new in OpenRAG v0.4
OpenRAG brings together the essential components developers need to build intelligent applications over their documents, combining Docling for document ingestion, OpenSearch for retrieval, and Langflow for workflow orchestration into a single integrated platform.
With OpenRAG v0.4, we focused on three priorities:
- Document Understanding
- Retrieval and Agent Orchestration
- Improved Reliability and Developer Experience
OpenRAG v0.4 improves how documents are ingested, processed, and prepared for retrieval. The platform continues to build on Docling’s structured document processing to convert complex documents such as PDFs, forms, and reports into structured, searchable knowledge. Improvements to ingestion validation, file handling, and document processing help ensure that uploaded documents are properly processed and retrievable, leading to higher-quality retrieval and better LLM responses.
Retrieval and Agent Orchestration
OpenRAG combines semantic search with orchestration workflows to support more advanced RAG and agent-based applications. OpenSearch powers vector and semantic retrieval, while Langflow enables developers to build and customize agent workflows, retrieval pipelines, and tool usage. Improvements in search behavior, filtering, and knowledge management help developers more easily organize and retrieve information across their document collections.
Improved Reliability and Developer Experience
A major focus of OpenRAG v0.4 was stability, installation, and overall developer experience. This release includes improvements to onboarding, configuration validation, container persistence, Windows and macOS compatibility, performance after upgrades, and overall system reliability. New storage connector Amazon S3 was also added, making it easier to integrate OpenRAG into real-world environments and production workflows.
Introducing IBM Elite Support for OpenRAG
As adoption of OpenRAG grows, many teams want the flexibility of open source combined with the reliability of enterprise support. IBM Elite Support for OpenRAG provides:
- Production support for OpenRAG deployments
- Direct access to maintainers and technical experts
- Guidance on architecture and scaling
- Faster issue resolution for production environments
This offering enables teams to deploy OpenRAG with confidence while maintaining the flexibility of open source.
If you are new to OpenRAG, check out the OpenRAG Quickstart to build an agentic RAG workflow in under 15 minutes or explore the OpenRAG GitHub repository.
If you are interested in learning more about IBM Elite Support for OpenRAG, check out the announcement or connect with a verified IBM Partner for more options.