DeepSeek is the name of a Chinese company currently shaking up the world of artificial intelligence (AI) because it claims to have successfully trained its AI models at a cost thousands of times lower using less powerful hardware due to US technology barriers. After the DeepSeek R1 was launched last December, they launched the Janus-Pro-7B model which can generate and understand the images it sees.
For your information, Janus-Pro-7B is also said to have higher capabilities than DALL-E by OpenAI and Stable Diffusion by Stable Diffusion. This image generation ability is matched with the ability to understand what is seen, allowing it to be used in robots that perform automation processes in factories. The latest model of Janus-Pro-7B can be accessed via the DeepSeek GitHub page today.
In addition, China's manufacturing strength lies in its lower costs at a larger scale that is difficult for other countries to match. Now, the same capabilities also seem to be being applied by China in the development of domestic AI. This is a major reason why the United States is increasingly restricting access to high-powered CPUs and GPUs to countries that are not close allies to ensure that their supremacy in AI is not threatened.
What are your thoughts on this news? Comment down below and stay tuned for more news like this at TechNave!
COMMENTS