Chinese AI firm DeepSeek released a new open-source model it said would compete with OpenAI’s and Google DeepMind’s advanced products, as China races to catch up to the US on AI.
Hangzhou-based DeepSeek took the tech world by storm last year by releasing a model that delivered near industry-leading performance at a fraction of the price, alleging that it made far more efficient use of chips than US champions.
Chinese AI developers still lack access to the world’s most advanced chips, but Nvidia CEO Jensen Huang recently suggested the country’s plentiful, cheap energy allows firms to use more low-efficiency chips to reach the same results.
