2024-02-25 03:01:47
2) What distinguishes the “H100” chip over others?
Generative AI applications learn to perform tasks, such as translating text, summarizing reports, or composing images, by training on huge amounts of pre-existing material. The more materials become available for these applications, the better they will become at skills such as recognizing human language or writing job application letters. They develop through trial and error, making billions of attempts until they are perfected, and in doing so they consume huge amounts of computing power.
NVIDIA says that the H100 chip is 4 times faster than the previous chip, the A100, in training so-called large linguistic models (LLM), and 30 times faster in responding to user commands. This feature can be very important for companies competing to train large language models to perform new tasks.
3) How did NVIDIA become a leader in artificial intelligence?
Nvidia, which is headquartered in Santa Clara, California, is the world’s most prominent company in the field of graphics chips, which are the chips responsible in computers for generating the images you see on the screen. The most powerful of these chips are those made with hundreds of cores or processing centers that run many processes and programs on the computer simultaneously, simulating complex physics such as shadows and reflections.
Goldman: Nvidia shares are the most influential in the world now
Since the beginning of this century, Nvidia engineers have realized that they can reconfigure graphics and image accelerators to work on other applications, by dividing tasks into smaller, simpler units and operations, and then working on them in parallel simultaneously. Just a decade ago, artificial intelligence researchers discovered that their research could finally turn into practical ideas using this type of chip.
1708843509
#NVIDIA #chip #H100 #jumped #artificial #intelligence #stocks