Chinese AI startup DeepSeek released an update to its logical reasoning model R1 on Thursday morning, intensifying competition with U.S. counterparts like OpenAI.
اضافة اعلان
The updated model, titled R1-0528, was made available on the developer platform Hugging Face, although no official announcement or documentation has yet been published by DeepSeek. The company has not released benchmark comparisons or detailed descriptions of the model's capabilities.
However, according to the LiveCodeBench leaderboard—a benchmark developed by researchers at UC Berkeley, MIT, and Cornell—the updated R1 model ranks just below OpenAI’s O4-mini and O3 models in code generation tasks, while outperforming Grok-3-mini from XAI and Qwen 3 from Alibaba.
According to Bloomberg, a DeepSeek representative described the release on a WeChat group as a "simple experimental update," encouraging users to begin testing the model.
Meanwhile, Alibaba also announced an updated version of its flagship Qwen AI model, reflecting the accelerating pace of innovation in China’s AI sector.
DeepSeek made headlines earlier this year with the launch of the original R1, challenging assumptions that U.S. export restrictions would stifle China’s AI progress. The model received widespread praise for achieving performance on par with top Western models—at significantly lower computational cost.
The release triggered a global tech selloff and renewed debate around whether cutting-edge AI requires massive capital and compute resources.
In response to DeepSeek’s rapid ascent, major Chinese tech firms like Alibaba and Tencent launched rival models claiming superiority. Meanwhile, OpenAI and Google revised their pricing strategies, with OpenAI introducing the lightweight o3 mini model to reduce compute requirements.
DeepSeek, headquartered in Shanghai, is widely expected to release its next-generation R2 model soon. According to sources cited by Reuters in March, the R2 release is scheduled for May. The company also introduced an enhanced multilingual version of its V3 model earlier this year.
— Agencies
