OpenResearcher

TIGER-AI-Lab's fully open deep-research recipe: 96K long-horizon trajectories (adopted by NVIDIA Nemotron), a 30B-A3B model hitting 54.8% BrowseComp-Plus, training code and eval harness.

976 93 Pythonupdated 1 months ago

training

View on GitHub Dispute this mapping →

Curator's take

The open counterpoint to closed Deep Research products — and the data is the crown jewel: 100+-turn research trajectories distilled from GPT-OSS-120B over a self-built 11B-token retriever corpus (no search-API bills at generation scale), good enough that NVIDIA folded it into Nemotron 3 Ultra. The 30B-A3B model beats GPT-4.1, Claude-Opus-4 and Gemini-2.5-Pro on BrowseComp-Plus. Reproducing anything is a real commitment: the setup assumes 8×A100, training lives in a Megatron-LM fork, and the local retriever needs Java + tevatron. ⚠ No LICENSE file in the repo — clarify terms before commercial use. Pick DeepDive for the KG-synthesis + multi-turn-RL recipe; OpenResearcher for large-scale SFT distillation with everything — data, model, eval — actually released.

Mapped by ShipWithAI editors · links verified

README.md

Adopted by NVIDIA's Nemotron family of models!

🤗 HuggingFace ｜ Slack | WeChat

📣 News

[2026.6.05] 🔥 OpenResearcher data has been adopted by Nemotron 3 Ultra!
[2026.3.25] 🚀 OpenResearcher hits #2 on Hugging Face Daily Papers!
[2026.3.24] OpenResearcher paper is now available, highlighting practical insights into deep research pipeline design. Already adopted by NVIDIA's Nemotron family of models!
[2026.2.25] Honored to be among the top 3 trending datasets on 🤗 Hugging Face — now 11K+ downloads!
[2026.2.18] The OpenResearcher training code is now available. Start training your own OpenResearcher!
[2026.2.14] Excited to have our OpenResearcher demo video. Dive in and unlock the power of Deep Research today!
[2026.2.12] Excited to see OpenResearcher powering deep research trajectory generation in NVIDIA’s NeMo Data Designer!
[2026.2.10] Our X post received 1.2K+ likes! Feel free to check out the post and join the discussion!

💥 Introduction

OpenResearcher is a fully ope

Continue your stack

What teams reach for next — and why each earns a place beside OpenResearcher. Ranked by curator confidence.

Alternative to See all + compare →

DeepDive

THUDM recipe for deep-search agents: synthesize hard multi-hop QA from knowledge-graph random walks, then multi-turn GRPO RL — DeepDive-32B hits 14.8% BrowseComp; data feeds GLM-4.5/4.6.

Why it fitsRival open recipes for training deep-research agents: DeepDive synthesizes hard QA from knowledge-graph walks and trains with multi-turn RL on slime (checkpoints still pending); OpenResearcher distills 96K long trajectories from GPT-OSS-120B and ships data, 30B model and eval framework complete.

Built with / on

vLLM

High-throughput, memory-efficient inference and serving engine for LLMs.

Why it fitsThe whole serving path runs on vLLM — the 30B-A3B deploys via bundled vLLM server scripts and the trajectory generation leans on vLLM's native browser-tool support for GPT-OSS.