In an exciting turn of events, China’s AI startup DeepSeek has launched its latest model, the DeepSeek-V3-0324. This bold move is aimed at taking on the big players in the U.S. AI industry. The new model brings significant upgrades to the V3 large language model, with a strong focus on enhancing reasoning and coding capabilities. According to a Reuters report, benchmark tests on the Hugging Face platform highlight these advancements.
The DeepSeek-V3-0324 is now available on Hugging Face, and it’s catching the eye of many in Silicon Valley. It’s said to match the performance of OpenAI’s ChatGPT but with a much lower investment. DeepSeek revealed that their V3 model was developed with less than $6 million worth of computing power using 2,000 Nvidia H800 chips. This is quite a contrast to the hefty investments made by leading U.S. tech companies in high-end chips and expansive data centers.
DeepSeek’s rise could mark a pivotal shift in the AI landscape, reminiscent of a “Sputnik moment” in the tech race between the U.S. and China. This development challenges the common belief that the U.S. holds a technological edge over China in AI advancements.
For those of us keeping an eye on the evolving tech rivalry, this is definitely a development to watch closely.
