Chinese AI startup DeepSeek has restored access to its API after nearly three weeks of service disruption caused by capacity limitations. The company resumed API credit top-ups on Tuesday, enabling developers to build apps and services using DeepSeek’s cloud-hosted AI. However, a company representative warned that server resources remain strained during peak daytime hours, which may impact performance.
DeepSeek gained significant attention earlier this year after launching its openly available R1 “reasoning” model. This AI model competes with some of OpenAI’s top offerings, matching or even surpassing their performance in certain areas. DeepSeek’s rapid rise has pushed OpenAI to reconsider its approach to open-sourcing AI technology and accelerating the release of select products.
The AI competition in China is intensifying, with several companies ramping up their efforts. On the same day DeepSeek resumed API top-ups, Alibaba introduced a preview of its latest reasoning AI model, QwQ-Max. The tech giant has announced plans to make this model open source, further increasing competition in the industry.
Despite facing infrastructure challenges, DeepSeek remains a key player in the AI market, particularly in the reasoning model segment. The temporary service pause highlights the increasing demand for its technology, but it also raises questions about the company’s ability to scale efficiently. As DeepSeek continues to develop its models and services, maintaining stable access to its API will be crucial for sustaining its growth and competing with major players like OpenAI and Alibaba.
