An Open-Source Alternative to OpenAI's Deep Research: Open Deep Research

After OpenAI introduced Deep Research, many open source versions comes up. Actually, there was already some built agents similar to deep research.

Huggingface published a blog post about that and they also run some agents for it.

Other great open implementations of Deep Research emerged from the community, specifically from

• dzhng,
• assafelovic(This even exists before Deep Research),
• nickscamara,
• jina-ai and
• mshumer.

Each of these implementations use different libraries for indexing data, browsing the web and querying LLMs.

In this article, I want to mentionNicolas Silberstein Camara for his great action against Openai Deep Research Premium. Here is the details and tutorial.

Open-Source Deep Research: Overview

Open Deep Research is anopen-source clone of OpenAI'sDeep Research experiment. Unlike OpenAI's proprietary model, this project leveragesFirecrawl’s extract + search technology combined with areasoning model to conduct deep research across the web.

Open-Source Deep Research: Key Features

• Firecrawl Extract + Search

  • Feeds real-time data to the AI viasearch.
  • Extractsstructured data from multiple websites.

• Next.js App Router

  • UsesReact Server Components (RSCs) for efficient rendering.
  • Supportsserver-side rendering for performance optimization.

• AI SDK Integration

  • Supports multipleLLM providers:
  • OpenAI (default:gpt-4o)
  • Anthropic, Cohere, DeepSeek, and more.

• Advanced UI Components

  • Styled withTailwind CSS.
  • Usesshadcn/ui withRadix UI for flexible component handling.

• Data Persistence

  • UsesVercel Postgres (Neon) for chat history and user data.
  • Stores files efficiently withVercel Blob.

• Authentication System

  • Implemented usingNextAuth.js for secure user login.

Open Deep Research: Installation Guide (Run Locally)

1. Clone the Repository

Open your terminal and run:

git clone https://github.com/nickscamara/open-deep-research.gitcdopen-deep-research

2. Install Dependencies

Installpnpm if not installed:

npminstall-g pnpm

Then install all dependencies:

pnpminstall

3. Set Up Environment Variables

You'll need to define environment variables in.env using the.env.example file as a reference.

To automatically configure environment variables:

vercelenvpull

Or manually create a.env file and include:

OPENAI_API_KEY=your_openai_api_keyFIRECRAWL_API_KEY=your_firecrawl_api_keyAUTH_SECRET=your_auth_secret

⚠️Do not commit the.env file to avoid exposing sensitive API keys.

4. Run Database Migrations

pnpm db:migrate

5. Start the Application

pnpm dev

Your app should now be running at:

🔗http://localhost:3000

Alternative Deployment: One-Click Deploy to Vercel

If you prefernot to run it locally, you candeploy to Vercel in one click:

Model Providers

By default, the project usesOpenAI's GPT-4o.

However, itsupports multiple LLM providers viaVercel's AI SDK, including:

• Anthropic (Claude)
• Cohere
• DeepSeek
• TogetherAI
• OpenRouter

Switching Models

Modify the.env file:

REASONING_MODEL=deepseek-reasonerBYPASS_JSON_VALIDATION=true

Adding Model Dependencies

If you want touse a model other than GPT-4o, install the respective dependency.

DeepSeek AI Model

pnpm add @ai-sdk/deepseek

TogetherAI Model

pnpm add @ai-sdk/togetherai

🚨Check TogetherAI rate limits:

🔗Rate Limit Info

Reasoning Model Configuration

This project includes areasoning model forstructured outputs such as research analysis, data extraction, and document summarization.

Key Notes

• GPT-4o, o1, o3-mini → Natively supportstructured JSON outputs.
• DeepSeek & TogetherAI → NeedBYPASS_JSON_VALIDATION=true.
• If no model is set, it defaults too1-mini.
• If an invalid model is chosen, it falls back too1-mini.

To useDeepSeek as the reasoning model, add this to.env:

REASONING_MODEL=deepseek-reasonerBYPASS_JSON_VALIDATION=true

Open-Source Deep Research: Conclusion

Open Deep Research byNicolas Silberstein Camara is a powerful,open-source alternative to OpenAI’sDeep Research.

It allows users toautonomously research the web, retrievestructured data, and leveragemultiple AI models.

By following the steps above, you can:
✅Run it locally

✅Deploy it to Vercel

✅Customize model providers

For the latest updates, visit theGitHub Repository.

Developer:Nicolas Silberstein Camara

GitHub Repository:Github

Demo:Live Demo