OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Raglight

Raglight is a lightweight, modular Python framework for building Retrieval-Augmented Generation systems with configurable language models, embedding providers, vector stores and MCP-connected external tools and data sources.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Raglight — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Raglight when a RAG application needs a modular Python architecture and the ability to change model, embedding or vector-store providers. Its MCP integration is relevant when the application must connect to external tools or data sources.

Who it is for

  • Python developers building document-grounded AI applications.
  • Teams comparing language-model, embedding and vector-store combinations.
  • Builders adding external tools or data sources to RAG workflows through MCP.

Documented uses

  • Ingest and index supported documents for retrieval.
  • Combine document retrieval with language-model generation.
  • Swap LLM, embedding and vector-store components.
  • Create standard or agentic RAG workflows.
  • Expose a RAG application through the documented REST API serving path.
Before you choose

Requirements & limits

Raglight is available as a public MIT-licensed repository. The documented package installation uses pip; provider-specific services and credentials may be required.

  • Provider choice: documented integrations include Ollama, Google Gemini, LM Studio, vLLM, OpenAI, Mistral and AWS Bedrock.
  • Storage choice: ChromaDB and Qdrant are documented as optional vector-store extras.
  • Workflow choice: standard RAG, agentic RAG, hybrid search and query reformulation are documented capabilities.
  • Tool access: MCP integration connects the application to external tools and data sources.
  • Runtime setup: LM Studio requires the chosen model to be loaded, and AWS Bedrock requires configured credentials.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Start a document chat with the CLI wizard

Set up an interactive RAGLight document-chat workflow through the documented command-line wizard.

Before you start

  • Python environment with the RAGLight package installed.
  • A local LLM service such as Ollama running.
  • Documents available at a local path.

Steps

  1. Install the base RAGLight library with pip.

  2. Ensure a local LLM service such as Ollama is running.

  3. Run the documented raglight chat command.

  4. Follow the wizard prompts to select the local data source and chat with the documents.

Expected outcome

An interactive command-line workflow for chatting with the selected documents.

Guide limitations

  • The documented quick-start prerequisite is a running local LLM service such as Ollama.
  • The wizard stages described here cover setup and document chat, not a separately verified deployment.
Documentation 1 ↗
Source-based practical guide

Install a selected vector-store integration

Prepare RAGLight with an optional vector-store backend.

Before you start

  • A Python environment with pip.
  • A choice between the documented ChromaDB or Qdrant integration.

Steps

  1. Install the base RAGLight package or select a vector-store extra.

  2. Install raglight[qdrant] for Qdrant or raglight[chroma] for ChromaDB.

  3. Install raglight[langfuse] as an additional extra when Langfuse tracing is required.

Expected outcome

The selected optional vector-store integration, with optional Langfuse support, is added to the Python environment.

Guide limitations

  • The documented ChromaDB extra requires a C++ compiler on Windows.
  • This recipe covers package setup only and does not establish a successful application run.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Retrieval-augmented applications

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
RaglightIngest and index supported documents for retrieval.Cloud, Self-hosted, ServerRaglight is available as a public MIT-licensed repository. The documented package installation uses pip; provider-specific services and credentials may be required.
LangsearchCrawl website or local data for a knowledge base.CLI, Self-hosted, Server
Hybrid Search RagCombine document retrieval and language-model generation in a configurable workflow.
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
4 October 2026
Original website title
Bessouat40/RAGLight · Official homepage title · retained in its source language
How we review sourcesSuggest a correction