OpenAI CEO Sam Altman calls DeepSeek’s R1 model “impressive”
Deepseek, which has garnered international attention, recently shared a paper revealing that training its DeepSeek-V3 model cost it $5.6 million and it needed lower-cost hardware.
CEO Sam Altman has called the R1 AI model from Chinese startup DeepSeek “impressive”.
"DeepSeek's r1 is an impressive model, particularly around what they're able to deliver for the price. We will obviously deliver much better models and also it's legit invigorating to have a new competitor! we will pull up some releases," said Altman on X.
"But mostly we are excited to continue to execute on our research roadmap and believe more compute is more important now than ever before to succeed at our mission," he added.
Deepseek, which has garnered international attention, recently shared a paper revealing that training its DeepSeek-V3 model cost it $5.6 million and it needed lower-cost hardware.
Launched last week, DeepSeek-R1 is said to be 20 to 50 times more cost-efficient than OpenAI's o1 model, depending on the task, according to a post on DeepSeek's official WeChat account.
DeepSeek has spent under $6 million to power its base AI model, a fraction of the hundreds of millions or even billions of dollars that US companies invest in their AI technologies, said the startup. It developed R1 using Nvidia's H800 chips, which cost much less than high-end computing infrastructure.
Tech investor Marc Andreessen has called the R1 model “AI’s Sputnik moment”.
The R1 model is said to demonstrate capabilities comparable to well-known systems such as OpenAI’s GPT-4, Meta’s Llama, and Google’s Gemini, but with significantly lower development costs.
Meanwhile, tech stocks plummeted on Monday as DeepSeek startled the AI market. AI leader Nvidia faced a record one-day market value loss of $593 billion, the largest single-day drop for any company on Wall Street.
Edited by Swetha Kannan
