This model was uploaded to HF almost 6 weeks ago, but the supporting paper came out only yesterday. Odd, but nevertheless it's interesting. They claim it has state-of-the-art performance, beating OpenAI, Gemini and Kimi's Research bot implementations. Would be good if there are actually performant, open weights, sovereign(-ish) deep research LLMs.
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 benchmark.