TL;DR

Ragflow now supports Google BigQuery as a data source and introduces a new Snowball stemmer (NLP technique) for improved language support, enhancing document ingestion and text processing capabilities.

New

• Google BigQuery Connector: Ingest and sync data from Google BigQuery.
• Enhanced Language Support: Added a Snowball stemmer supporting 16 languages, including Dutch, for better text analysis.
• Document Ingestion API: New endpoint for customized document processing pipelines.

Fixes Worth Knowing

• Resolved crashes related to LM-Studio model names and MCP server communication.
• Fixed issues with Docling parser dropping formulas and metadata persistence.
• Corrected problems with bulk linking files, tag weights, and mobile UI layout.
• Addressed issues with chat message deletion and avatar uploads.

Before You Upgrade

If you are using the Ollama provider, ensure your API key is correctly configured, as a previous error has been resolved.

TL;DR

Ragflow now supports LangChain (a framework for building LLM applications) expression language, enabling more complex and flexible retrieval-augmented generation pipelines.

New

• LangChain Expression Language Support: Build sophisticated RAG pipelines using LangChain’s declarative syntax.

Fixes Worth Knowing

• Improved error messaging for invalid data sources.
• Resolved an issue where certain document loaders failed silently.

Before You Upgrade

• Update your LangChain version to at least 2.10 to take full advantage of the new expression language support.