# RAGLight

> RAGLight is a modular framework for Retrieval-Augmented Generation (RAG). It makes it easy to plug in different LLMs, embeddings, and vector stores, and now includes seamless MCP integration to connec

- **URL**: https://www.freshcrate.ai/projects/RAGLight
- **Author**: Bessouat40
- **Category**: MCP Servers
- **Latest version**: `3.4.7` (2026-03-24)
- **License**: MIT
- **Source**: https://github.com/Bessouat40/RAGLight
- **Homepage**: https://raglight.mintlify.app/
- **Language**: Python
- **GitHub**: 658 stars, 100 forks
- **Registry**: github
- **Tags**: `agentic-ai`, `agentic-rag`, `agentic-workflow`, `artificial-intelligence`, `data-science`, `framework`, `huggingface`, `lmstudio`, `python`

## Description

RAGLight is a modular framework for Retrieval-Augmented Generation (RAG). It makes it easy to plug in different LLMs, embeddings, and vector stores, and now includes seamless MCP integration to connect external tools and data sources.

## Recent releases

| Version | Date | Urgency | Changes |
| --- | --- | --- | --- |
| `3.4.7` | 2026-03-24 | Medium | ## What's Changed * Add streaming support for all LLM providers by @Bessouat40 in https://github.com/Bessouat40/RAGLight/pull/138 * Add Bedrock Provider * Add tests for RAG pipeline, streaming, document processors, and scrapper by @Bessouat40 in https://github.com/Bessouat40/RAGLight/pull/139 * Add qdrant vector store + add some tests by @Bessouat40 in https://github.com/Bessouat40/RAGLight/pull/141 * Fix Qdrant hybrid search by @Bessouat40 in https://github.com/Bessouat40/RAGLight/pull/142 |
| `3.2.0` | 2026-03-12 | Low | ## What's Changed * Improve/UI by @Bessouat40 in https://github.com/Bessouat40/RAGLight/pull/131 * add reformulation by @Bessouat40 in https://github.com/Bessouat40/RAGLight/pull/133 * improve history management by @Bessouat40 in https://github.com/Bessouat40/RAGLight/pull/134   **Full Changelog**: https://github.com/Bessouat40/RAGLight/compare/3.1.1...3.2.0 |
| `3.1.1` | 2026-03-05 | Low | Fix CLI bug |
| `3.1.0` | 2026-03-05 | Low | ## Added  - `raglight serve --ui` — launches a Streamlit chat interface alongside the REST API   - `--ui-port` option to configure the Streamlit port (default: `8501`)   - Chat tab with full conversation history and markdown rendering   - Upload tab to ingest files or a server-side directory directly from the browser   - Both processes share the same configuration and shut down together on Ctrl+C |
| `3.0.0` | 2026-03-04 | Low | # New Features                                                                                                                                                                                                                                                                                                                                                                                                             ## REST API Server (https://github.com/Bessouat40/RAGLight/pull/126) |
| `2.10.1` | 2026-01-12 | Low | # Update dependencies  Remove smolagents dependency. |
| `2.10.0` | 2026-01-11 | Low | # v2.10.0  ## What's Changed  - **[BREAKING]** Remove old RAT feature - **[FEATURE]** You can now add one or more github repositories as knowledge source when you use RAGLight's CLI. You can now discuss with your favorites repo using the framework. |
| `2.9.0` | 2026-01-09 | Low | ## Dependencies update  Update dependencies to avoid vulnerabilities.  ## Reranking on classic RAG  Improve reranking feature on RAG feature.  ## Memory  Add memory use for RAG feature.  ## New PDF parser  Possibility to use a new pdf parser based on a vlm that can deals with images inside pdf files when you index documents.  ## Refactor Agentic RAG Feature  Refactor of Agentic RAG Feature : use langchain framework instead of smolagents one. |
| `2.8.1` | 2025-12-04 | Low | Fix : modify RAG class for better context understanding. |
| `2.8.0` | 2025-12-04 | Low | # Custom Processors & VLM-Powered PDF Parsing  ## New Feature: Override Vector Store Processors  You can now override RAGLight's default document processors when indexing your data. This makes the ingestion pipeline fully customizable.  ```python custom_processors = {     "pdf": VlmPDFProcessor(vlm),     "py": MyCustomPythonProcessor(), } ```  This allows you to plug in your own logic for any file extension and adapt RAGLight to advanced or domain-specific workflows.  ---  ## Ne |

## Dependency audit

- **Score**: 66/100
- **Total deps**: 27
- **Resolved**: 16
- **Unresolved**: 11
- **License conflicts**: 0
- **Warnings**: 13
- **Scanned**: 2026-05-04

## Citation

- HTML: https://www.freshcrate.ai/projects/RAGLight
- Markdown: https://www.freshcrate.ai/projects/RAGLight.md
- Dependencies JSON: https://www.freshcrate.ai/api/projects/RAGLight/deps

_Generated by freshcrate.ai. Indexes github releases for AI-agent ecosystem packages._