OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Open Deep Wide Research

Open Deep Wide Research is an open-source agentic RAG application that answers research questions using selected data sources, with controls for retrieval depth and information breadth.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Open Deep Wide Research — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when building a research or RAG workflow where source selection and retrieval scope need to be configurable, and you can provide the required model and search-provider credentials.

Who it is for

  • Developers building RAG chatbots or research assistants
  • Teams connecting internal knowledge bases or specialized APIs to agentic research workflows
  • Builders who want to adjust retrieval scope for customer-service, market-research or analytics use cases

Documented uses

  • Connect custom information sources, including internal knowledge bases and specialized APIs.
  • Adjust research depth and breadth to shape a response for a particular query.
  • Build a web-interface research assistant or use the backend-only deployment option described in the README.
  • Configure MCP-compatible data sources and tools, or choose among the listed model and search-provider options.
Before you choose

Requirements & limits

The repository is public and licensed under Apache-2.0. The documented setup uses provider API keys; the README lists OpenRouter as required and Exa or Tavily as at least one required search provider.

  • Backend setup requires Python 3.9 or later, an OpenRouter API key, and at least one Exa or Tavily key.
  • The full web interface additionally uses Node.js 18 or later.
  • The README describes MCP support, Tavily and Exa search providers, and model options including OpenAI, Claude and private LLMs.
  • The project documents cost as tied to depth and breadth settings, while its sample timings and amounts are scenario examples.
  • Deployment choices include backend-only API use or a full frontend-and-backend setup.

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 the backend environment

Prepare the documented Python backend environment and configure its required provider keys.

Before you start

  • Python 3.9 or later
  • An OpenRouter API key
  • An Exa or Tavily API key

Steps

  1. Copy deep_wide_research/env.example to deep_wide_research/.env.

  2. Add the OpenRouter key and at least one Exa or Tavily key to the environment file.

  3. Enter the deep_wide_research directory, create and activate a Python virtual environment, then install requirements.txt.

Expected outcome

The backend environment is configured and its Python dependencies are installed; the README's next documented step is starting the backend with python main.py.

Guide limitations

  • This recipe covers setup only; it does not verify that the backend starts or completes a research request.
  • Provider credentials are required for the documented backend configuration.
Documentation 1 ↗
Source-based practical guide

Prepare the web interface for local development

Install frontend dependencies and start the documented development server.

Before you start

  • Node.js 18 or later
  • The backend setup and provider credentials described in the project README

Steps

  1. Copy chat_interface/env.example to chat_interface/.env.local.

  2. Enter the chat_interface directory and run npm install.

  3. Run npm run dev and open http://localhost:3000.

Expected outcome

The README directs users to the local web interface at http://localhost:3000.

Guide limitations

  • The README does not specify the contents required in the frontend environment file.
  • These are documented setup instructions, not a verification that the interface or a research workflow runs successfully.
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
Open Deep Wide ResearchConnect custom information sources, including internal knowledge bases and specialized APIs.WebThe repository is public and licensed under Apache-2.0. The documented setup uses provider API keys; the README lists OpenRouter as required and Exa or Tavily as at least one required search provider.
VectaurantFind menu options by describing preferences, such as asking what is vegan or spicy.CLI, Docker, Linux, macOS
Aml Multiagent RagIngest PDF regulations, extract and chunk their text, and build a searchable vector collection.Docker, Self-hostedThe repository is public and specifies the MIT license. The documented setup is self-hosted and requires Python 3.11 or later, Docker Desktop and an OpenAI API key.
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
puppyone-ai/DeepWideResearch · Official homepage title · retained in its source language
How we review sourcesSuggest a correction