On the afternoon of November 3, DeepSeek experienced yet another large-scale service disruption, with both its web interface and API endpoints failing simultaneously, marking at least the 19th major outage the platform has suffered this year.
Starting at 16:58 on November 3, DeepSeek's official status page confirmed partial service interruptions, affecting the Chat Service, the DeepSeek V4 Pro API, and the DeepSeek V4.1 Flash API. When users attempted to use the chat function, the interface displayed a message reading "Server is busy, please try again later." The hashtag #DeepSeekDown quickly surged onto social media trending lists.
DeepSeek later updated its status page stating that fixes had been implemented for the web and API service disruptions and that the situation had entered a monitoring phase. However, as of the latest status update, the V4 Pro and V4.1 Flash API services as well as the chat service were still experiencing degraded performance and had not yet fully returned to normal.
Scope of the outage: both regular users and developers affected
This disruption affected two major groups: end users and developers. On the end-user side, neither the web interface nor the mobile app could properly return replies. On the developer side, the simultaneous failure of the DeepSeek V4 Pro API and V4.1 Flash API services meant that enterprises and developers who had integrated the model into their own products were unable to make API calls, dealing a direct blow to downstream commercial applications reliant on the service.
On social platforms, users reported that affected scenarios spanned a wide range of daily work needs, including article editing, code writing, and academic paper composition.
Analysts pointed out that the core cause of the frequent service disruptions lies in a severe imbalance between computing power supply and demand. As user numbers and API call volumes continue to grow, DeepSeek still faces considerable pressure in terms of infrastructure capacity and stability assurance.
At least the 19th time this year: outages cluster since October
Statistics show that this is at least the 19th large-scale anomaly DeepSeek has experienced this year, with failures becoming especially密集 since October.
On October 2, web and API services were disrupted for approximately two hours, involving the V4 Pro API, the V4.1 Flash API, and the chat service. On October 8, the platform experienced degraded service performance lasting about three minutes. On October 12, DeepSeek suffered two separate failures in a single day, one of which simultaneously affected the chat service and the V4.1 Flash API. On October 15, the platform once again encountered a large-scale disruption, with both the web interface and API endpoints failing simultaneously, and the outage persisted for nearly three hours without recovery.
Comments