Chinese AI startup DeepSeek is working on an artificial intelligence (AI) model featuring more sophisticated agent functions to rival the likes of OpenAI, according to a Bloomberg report.
The startup is developing an AI model that can perform multi-step tasks on a user’s behalf with minimal input, the report said, quoting people familiar with the matter.
The system is designed to adapt and get better through its past actions, the report noted, adding that DeepSeek founder, Liang Wenfeng, is pressing the team to launch the new software in the last quarter of this year.
YourStory could not independently verify the development.
In January, DeepSeek disrupted the global tech industry with the release of its R1 model, a powerful competitor that emulates human reasoning. The model reportedly cost only a few million dollars to develop. However, since then, DeepSeek has released only minor updates, while rivals in both the United States and China have introduced numerous new models.
Recently, major players such as OpenAI, Anthropic, and Microsoft have launched their own agent software to assist users with both personal and professional tasks. China-founded startup Manus AI has drawn global attention with its general-purpose AI agent.