OpenBMB/UltraRAG

📋 What's Changed

• 1. Add an end-to-end memory system with persistent user memory, project memory retrieval, and a dedicated memory-aware RAG demo. by @mssssss123 in #379
• 2. Add SQLite-backed authentication, persistent chat sessions, nickname/model settings management, and knowledge base visibility controls. by @mssssss123 in #379
• 3. Refactor the frontend with React/Vite, add the Explore section, and improve mobile responsiveness and source-detail interactions. by @mssssss123 in #379
• 4. Fix async retriever initialization by awaiting retriever_init(...) during startup. by @xhd0728 in #385
• 5. Fix builder compatibility issues caused by FastMCP 3.x API changes, and harden demo runtime handling around closed-event-loop cleanup noise and backend overrides. by @xhd0728 in #388
• 6. Add CI smoke checks for the SayHello pipeline and frontend build, then refine the workflow tooling and installation examples. by @xhd0728 in #389 and #390
• 7. Add a structured troubleshooting guide for UltraRAG workflows. by @onestardao in #352
• 8. Update README publications and model references. by @xhd0728 in #359

📋 What's Changed

• 1. Support web search functionality for the demo environment. by @xhd0728 in #285
• 2. Support .doc and .wps file parsing. by @xhd0728 in #293
• 3. Implement Case Study Viewer launch functionality. by @mssssss123 in #313
• 4. Add export functionality for chat messages in DOCX format. by @mssssss123 in #323
• 5. Simplify the parameter display toggle for a cleaner interface. by @mssssss123 in #282
• 6. Add user confirmation for interrupting chat generation and streamline the creation of new chat sessions. by @mssssss123 in #305
• 7. Implement automatic citation renumbering functionality. by @mssssss123 in #314
• 8. Enhance chat text actions with download capabilities and sort used/unused sources by numeric citation ID. by @mssssss123 in #315
• + 1 more

📦 Highlights

• WYSIWYG Pipeline Builder: From logic to prototype in seconds. Our dual-mode builder (Canvas + Code) syncs in real-time. Click "Build" and your static logic instantly becomes an interactive UI. No more boilerplate code!
• Pixel-Level "White-Box" Visualization: Stop guessing. The "Show Thinking" panel visualizes the entire inference trajectory—loops, branches, and tool calls. Debug bad cases instantly by comparing retrieval chunks vs. model hallucinations.
• Built-in AI Developer Assistant: Stuck on config? The embedded AI Assistant knows the framework inside out. Just use natural language to generate Pipeline configurations, optimize Prompts, or explain parameters.
• DeepResearch Engine: Powered by AgentCPM-Report , it supports "Writing-as-Reasoning." The system dynamically plans, retrieves, and deepens content to generate professional, cited reports automatically.

📋 What's Changed

• Demo Mode & Streaming:
• Enable demo-mode compatibility for retriever server and Milvus backend (#145) @mssssss123
• Enable streaming generation in generation server to support demo mode (#147) @mssssss123
• Full stack support (Frontend, Backend, Client) for Demo; Upgrade CLI to v3 (#149) @mssssss123
• Support displaying "thinking process" for demo (#151) @mssssss123
• Support customizable chunk params, optimize corpus server, and add thread cleanup (#167) @mssssss123
• Knowledge Base (KB):
• Support knowledge management for demo (#160) @mssssss123
• + 76 more

📋 What's Changed

• 1. Fixed index reference for image paths in generation server. by @xhd0728 #146
• 2. Reimplemented the search-o1 pipeline by @mssssss123 @lifelsl #158

📋 What's Changed

• 1. Added full deployment support for the retriever server. by @mssssss123 #136
• 2. Updated the UltraRAG demo front-end interface. by @mssssss123 #139
• 3. Optimized GPU/CPU configuration logic for the retriever server. by @mssssss123 #142
• 4. Introduced new ToolCall and PipelineCall functionality. by @hm1229 @mssssss123 #143

📋 What's Changed

• 4. Fixed the Search-o1 Pipeline: Resolved critical issues in the iterative search-generation loop, restoring stable functionality for the Search-o1 pipeline. by @mssssss123 #123
• 5. Fixed Accuracy Metric Calculation Bug: Corrected inaccuracies in the acc metric computation, ensuring more reliable and consistent evaluation results across benchmarks. by @mssssss123 #127
• 6. Added Script for Saving Retrieval Results: Introduced a new script to persist retrieved content, improving debugging workflows, visualization, and error case analysis. by @mssssss123 #129
• 7. Fixed Bugs in the WebNote Pipeline: Addressed multiple issues in the WebNote pipeline to improve robustness when processing webpage-like or long-form content. by @mssssss123 #128
• 9. Added Progress Bar for Document Parsing: Document parsing now displays a progress bar, providing clearer feedback during long document processing. by @mssssss123 #133
• 10. Updated README with Complete Docker Usage Guide: The README has been updated with a comprehensive Docker usage section, making deployment easier and more accessible. by @xhd0728 #131

📦 Highlights

• 1. Comprehensive Multimodal Upgrade: Both the Retriever and Generation Servers now support multimodal inputs, enabling a complete end-to-end multimodal workflow from retrieval to generation.

📋 What's Changed

• Corpus Server supports plain text extraction from .txt, .md, .pdf, .xps, .oxps, .epub, .mobi, and .fb2 files. @mssssss123
• Corpus Server adds simple per-page image conversion for .pdf files. @mssssss123
• Corpus Server integrates MinerU for high-precision PDF parsing. @mssssss123
• Corpus Server introduces a new chunking strategy supporting token-level (word/character) segmentation. @mssssss123
• Corpus Server supports sentence-level chunking. @mssssss123
• Corpus Server supports customizable chunking rules (default rule recognizes Markdown sections; other rules can be extended via config files). @mssssss123
• Retriever Server supports three retrieval engines: Infinity, Sentence-Transformers, and OpenAI. @mssssss123
• Retriever Server supports multimodal retrieval. @mssssss123
• + 14 more

📋 What’s Changed

• MCP Architecture & Server–Client Redesign
• Unified MCP Server Architecture: All RAG components (Benchmark, Corpus, Retriever, Generation, Evaluation, Router, Note) are now implemented as standalone MCP Servers, enabling modular development and clean extensibility.
• by @hm1229 @xhd0728
• Full MCP Client Pipeline Engine: Introduced a new pipeline executor supporting loop, branch, router, multi-server aliases, early stopping, and nested pipelines, making complex iterative workflows natively supported.
• by @hm1229 @xhd0728
• Corpus Server
• Added unified file-reading support (doc/docx/pdf/md/txt) and multiple chunking strategies (static/semantic/LLM).
• by @mssssss123
• + 37 more