ISLAMABAD: Chinese AI startup DeepSeek announced an update to its R1 reasoning model on Thursday, signaling intensified competition with US counterparts such as OpenAI. The upgraded version, R1-0528, was quietly launched on the developer platform Hugging Face without a formal public announcement or detailed technical description. Despite this, independent benchmarks like the LiveCodeBench leaderboard—developed by researchers from UC Berkeley, MIT, and Cornell—placed DeepSeek’s updated R1 model just behind OpenAI’s advanced o4 mini and o3 models in code generation, and ahead of xAI’s Grok 3 mini and Alibaba’s Qwen 3 models.
DeepSeek’s Rapid Growth in AI Landscape
Earlier this year, DeepSeek disrupted expectations by releasing AI models on par with top US offerings at a fraction of the cost, challenging the notion that AI development requires massive computing resources and investments. The January launch of R1 sent ripples through global tech markets, particularly outside China. Since then, other Chinese tech giants like Alibaba and Tencent have released their own models, aiming to surpass DeepSeek’s capabilities. Meanwhile, US firms like Google and OpenAI have introduced competitive pricing strategies and scaled-down versions of their models to maintain market share.
Looking Ahead: R2 and Future Developments
DeepSeek is widely expected to release R2, the successor to R1, which was initially planned for May according to earlier reports. In March, the company also updated its V3 large language model, underscoring its ongoing commitment to advancing AI technology. As DeepSeek continues to push boundaries, the race between Chinese and American AI companies grows fiercer, promising rapid innovation in the global AI landscape.
