TIGER-AI-Lab Open-Sources OpenResearcher Agent Pipeline

TIGER-AI-Lab released OpenResearcher, a fully open pipeline for training deep research agents, with 96K web-research trajectories and 54.8% on BrowseComp-Plus.

by HowAIWorks Team
On this page

Introduction

TIGER-AI-Lab has released OpenResearcher, a fully open pipeline for training deep research agents — AI systems that answer complex questions by searching, browsing, and reasoning across many steps rather than answering from memory. The release, described in the paper "A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis", goes well beyond a single model checkpoint: it publishes the training recipe, a dataset of roughly 96,000 research trajectories, training code, an evaluation framework, and an offline search environment for reproducing the whole pipeline end to end.

The headline number is the scale of the training data itself: TIGER-AI-Lab labels the release 96K long-horizon research trajectories (the paper's precise count is just over 97,000 sequences), many containing more than 100 tool calls apiece, built specifically to teach a model how to research rather than what to answer.

What a research trajectory teaches the model

Most instruction-tuned models answer a question directly from parameters learned during pretraining. OpenResearcher instead trains agents to work through a research loop:

  • Search for relevant documents
  • Open promising pages
  • Find the specific fragments that answer the question
  • Gather evidence across multiple sources
  • Reason over several steps before committing to an answer
  • Produce a final answer grounded in what was retrieved

Each of these trajectories is a full transcript of an AI agent working through this loop on a real research question, generated by using GPT-OSS-120B as a teacher model to explore an offline collection of 15 million documents — roughly 11 billion tokens — through exactly those three tools — search, open, and find.

Why an offline search environment

Training research agents against live web search APIs is slow, costly, and non-reproducible — results change over time and rate limits throttle large-scale trajectory generation. OpenResearcher sidesteps this by separating dataset construction from retrieval: the entire pipeline runs against a static, offline document index, so generating trajectories, fine-tuning on them, and evaluating the result are all cheap and deterministic. That same offline environment lets other researchers rerun the exact experiments and test their own filtering strategies or retrieval methods without needing API budget.

Results

The team used the trajectory dataset to fine-tune OpenResearcher-30B-A3B, a mixture-of-experts model with 30 billion total parameters and about 3 billion active per forward pass (the "A3B" in the name), via supervised learning. On BrowseComp-Plus, a benchmark for deep research agents, the fine-tuned model reached 54.8% accuracy — a 34.0-point improvement over its base model — while holding up on related benchmarks, including BrowseComp, GAIA, and xbench-DeepResearch.

What's released

Everything is public:

  • OpenResearcher-Dataset — the ~96K research trajectories, on Hugging Face
  • OpenResearcher-Corpus — the offline 15-million-document, ~11-billion-token collection used for retrieval
  • Training code and the distillation recipe used to generate trajectories from GPT-OSS-120B
  • Evaluation framework and logs for reproducing the benchmark results
  • Trained model weights for the 30B-A3B checkpoint

The repository is available at github.com/TIGER-AI-Lab/OpenResearcher.

Conclusion

OpenResearcher's contribution isn't a single leaderboard number — it's making the entire recipe for training a deep research agent reproducible, from raw document corpus through trajectory synthesis to the fine-tuned model. That matters for a category where most strong systems (commercial deep-research products from major labs) keep their training data and search infrastructure closed. By publishing the data, the offline environment, and the code together, TIGER-AI-Lab gives other teams a concrete baseline to fine-tune, extend, or audit rather than a black box to reverse-engineer.

Sources

Frequently Asked Questions

OpenResearcher is a fully open pipeline from TIGER-AI-Lab for training deep research AI agents. It includes the training recipe, a dataset of roughly 97,000 research trajectories, training code, an evaluation framework, and an offline web-research environment, not just a model checkpoint.
TIGER-AI-Lab describes the released dataset as 96K deep research trajectories; the accompanying paper puts the precise count at just over 97,000 sequences. Many trajectories include more than 100 tool calls each, generated by having GPT-OSS-120B research questions against an offline corpus of 15 million documents (about 11 billion tokens).
A 30B-parameter mixture-of-experts model, OpenResearcher-30B-A3B (30 billion total parameters, about 3 billion active per forward pass), fine-tuned on the OpenResearcher trajectories scores 54.8% on BrowseComp-Plus, a 34.0-point improvement over its base model, while holding up on related benchmarks like BrowseComp, GAIA, and xbench-DeepResearch.
Training against live search APIs is expensive, non-reproducible, and rate-limited. OpenResearcher instead indexes a static 15-million-document collection and exposes it through three tools — search, open, and find — so trajectory generation and evaluation are cheap, deterministic, and repeatable.
Everything is public on GitHub at TIGER-AI-Lab/OpenResearcher and on Hugging Face, including the OpenResearcher-Dataset, OpenResearcher-Corpus, evaluation logs, and the trained model weights.
A deep research agent is an AI system that answers open-ended questions by autonomously searching the web, opening and reading pages, extracting relevant evidence, and reasoning across multiple steps, instead of generating an answer purely from what it memorized during pretraining.

Continue Your AI Journey

Explore our glossary and model catalog to deepen your understanding.