The Year in AI Papers
The year/Independent research

Paper 2510.24701

Tongyi DeepResearch Technical Report

Tongyi DeepResearch Team, Baixuan Li, Bo Zhang, Dingchu Zhang, Fei Huang, Guangyu Li, Guoxin Chen, Huifeng Yin, Jialong Wu, Jingren Zhou, Kuan Li, Liangcai Su, et al.

Published
Oct 2025
Research lab
Independent
Citations
18
GitHub
Not linked

01 In brief

Summary

Tongyi DeepResearch is an open-source agentic large language model designed for long-horizon, deep information-seeking research tasks.

It uses an end-to-end training framework combining agentic mid-training (two-stage continual pre-training) and agentic post-training (SFT and reinforcement learning), supported by a fully automated synthetic data pipeline and stage-specific environments (prior world, simulated, real-world).

The model, with 30.5B total parameters and 3.3B activated per token, achieves state-of-the-art results on benchmarks including Humanity's Last Exam (32.9), BrowseComp (43.4), BrowseComp-ZH (46.7), WebWalkerQA (72.2), GAIA (70.9), xbench-DeepSearch (75.0), FRAMES (90.6), and xbench-DeepSearch-2510 (55.0), outperforming strong baselines like OpenAI-o3 and DeepSeek-V3.1.

A Heavy Mode with parallel research and synthesis further improves scores (HLE 38.3, BrowseComp 58.3, BrowseComp-ZH 58.1).

The model also shows gains on general benchmarks (AIME25, HMMT25, SimpleQA).

The authors release the model, framework, and solutions to empower community research.

02 From the paper

Abstract

We present Tongyi DeepResearch, an agentic large language model, which is specifically designed for long-horizon, deep information-seeking research tasks. To incentivize autonomous deep research agency, Tongyi DeepResearch is developed through an end-to-end training framework that combines agentic mid-training and agentic post-training, enabling scalable reasoning and information seeking across complex tasks. We design a highly scalable data synthesis pipeline that is fully automatic, without relying on costly human annotation, and empowers all training stages. By constructing customized environments for each stage, our system enables stable and consistent interactions throughout. Tongyi DeepResearch, featuring 30.5 billion total parameters, with only 3.3 billion activated per token, achieves state-of-the-art performance across a range of agentic deep research benchmarks, including Humanity's Last Exam, BrowseComp, BrowseComp-ZH, WebWalkerQA, xbench-DeepSearch, FRAMES and xbench-DeepSearch-2510. We open-source the model, framework, and complete solutions to empower the community.