Universal Deep Research: Bring Your Own Model and Strategy

Peter Belcak, Pavlo Molchanov
8/29/2025
cs.AI

No official code found. Best implementation: NVlabs/UniversalDeepResearch (467 stars, last commit 2025-08-26). 1 more below.

Abstract

Deep research tools are among the most impactful and most commonly encountered agentic systems today. We observe, however, that each deep research agent introduced so far is hard-coded to carry out a particular research strategy using a fixed choice of tools. We introduce Universal Deep Research (UDR), a generalist agentic system that wraps around any language model and enables the user to create, edit, and refine their own entirely custom deep research strategies without any need for additional training or finetuning. To showcase the generality of our system, we equip UDR with example minimal, expansive, and intensive research strategies, and provide a user interface to facilitate experimentation with the system.

View on arXivView PDF

Code Implementations(2)

davidkimai/universal-deep-researchCommunity71%
MIT

A minimal practical implementation of NVIDIA's Universal Deep Research system. https://arxiv.org/abs/2509.00244

41PythonSep 14, 20251 years agoMIT
NVlabs/UniversalDeepResearchCommunity69%

Code to accompany the Universal Deep Research paper (https://arxiv.org/abs/2509.00244)

46774CSS, ShellJul 11, 20251 years agoNOASSERTION
deep-researchdeep-research-agentinstruction-following

Similar Papers

CASCADE: Cumulative Agentic Skill Creation through Autonomous Development and Evolution

86% match2025

URSA: The Universal Research and Scientific Agent

84% match1 repo712025

Deep Research of Deep Research: From Transformer to Agent, From AI to AI for Science

84% match2026

Exploring Large Language Model based Intelligent Agents: Definitions, Methods, and Prospects

84% match1 repo1.9k2024

Tongyi DeepResearch Technical Report

83% match2025

OS-Copilot: Towards Generalist Computer Agents with Self-Improvement

83% match1 repo1.8k2024

Cite this paper

@article{belcak2025universal,
  title  = {Universal Deep Research: Bring Your Own Model and Strategy},
  author = {Peter Belcak and Pavlo Molchanov},
  year   = {2025},
  eprint = {2509.00244},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.00244v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.00244.svg)](https://codeofpaper.com/paper/2509.00244)

Discussion