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

Dewy

Dewy is an open-source Python knowledge-management project for AI applications. It parses, chunks, summarizes, and indexes document content, then provides semantic retrieval for RAG applications and AI agents.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Dewy — 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 Dewy when an AI application needs a document-processing and semantic-retrieval layer between its source documents and the model prompts it constructs.

Who it is for

  • Developers building retrieval-augmented generation applications
  • Teams adding document-based knowledge retrieval to AI agents
  • Python and JavaScript/TypeScript developers integrating a retrieval layer into an AI workflow

Documented uses

  • Extract and prepare document content for retrieval.
  • Index document data and embeddings for semantic search.
  • Retrieve relevant text chunks to add as context to an LLM prompt.
Before you choose

Requirements & limits

The project is publicly available on GitHub under the Apache-2.0 license. The README documents local installation and configuration.

  • The documented storage setup uses a vector database; pgvector is the README’s example.
  • A local setup requires an OpenAI API key and database configuration.
  • The repository documents Python service use and Python and TypeScript/JavaScript clients.
  • The roadmap presents additional formats, integrations, multimodal search, and ranking options as planned work rather than established current features.

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

Set up Dewy locally

Install and configure the Dewy service, then open its admin console.

Before you start

  • A local Python environment
  • An OpenAI API key
  • A PostgreSQL connection; the README documents pgvector for storing document metadata and embeddings

Steps

  1. Optionally start a pgvector instance with the documented Docker command, or prepare an existing pgvector database.

  2. Install Dewy with the command below.

    Documented command
    pip install dewy

  3. Set ENVIRONMENT=LOCAL, DB, and OPENAI_API_KEY in an .env file or the process environment.

  4. Run dewy, then open http://localhost:8000/admin.

Expected outcome

The README describes an admin console for creating collections, loading documents, and running test queries.

Guide limitations

  • The guide covers local service setup and access to the console; it does not document a complete application deployment.
Documentation 1 ↗
Source-based practical guide

Add documents and retrieve chunks with the JavaScript client

Use the documented TypeScript/JavaScript client flow to add a document and request retrieval context.

Before you start

  • A running Dewy instance
  • The dewy-ts package
  • A collection ID and a document URL

Steps

  1. Install the client with the command below.

    Documented command
    npm install dewy-ts

  2. Import Dewy from dewy_ts and create a client instance.

  3. Call dewy.kb.addDocument with the collection ID and document URL.

  4. Call dewy.kb.retrieveChunks with the collection ID, a search query, and the requested number of chunks.

Expected outcome

The documented retrieval call returns chunks that can be incorporated into an LLM prompt.

Guide limitations

  • The example assumes an existing Dewy instance and collection.
  • The README does not establish support for every document format.
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
DewyExtract and prepare document content for retrieval.CloudThe project is publicly available on GitHub under the Apache-2.0 license. The README documents local installation and configuration.
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
18 October 2026
Original website title
DewyKB/dewy · Official homepage title · retained in its source language
How we review sourcesSuggest a correction