The most powerful on the planet." Nvidia unveils new core for artificial intelligence companies in the artificial intelligence market. It is especially worth noting the new flagship core called Blackwell (B200).
New Power
- Blackwell cores are named after David Harold Blackwell, the first African-American scientist elected to the US National Academy of Sciences.
- NVIDIA claims that Blackwell is the most powerful core on the planet. They are 30 times faster than the H100 and consume 25% less energy.
- This core provides a significant performance boost for AI companies. Its speed is 20 petaflops compared to 4 petaflops of the H100.
- This high speed is made possible by the 208 billion transistors in the Blackwell cores compared to the 80 billion in the H100.
- NVIDIA has combined two large chips in these cores, which are capable of exchanging data at speeds of up to 10 terabytes per second.
- Blackwell will form the basis of future products for leading data center operators such as Amazon, Microsoft, Google and Oracle.
Glossary
- Blackwell is NVIDIA's flagship core, which is the most powerful on the planet
- H100 is a previous NVIDIA core, inferior in performance to Blackwell.
Link
In the press release of the NVIDIA company it is noted , that received high praise and reaction from seven CEOs of companies with a total value of trillions of dollars, which shows how important the operation of artificial intelligence on NVIDIA chips is by many leading firms.
Among them are OpenAI CEO Sam Altman, General Microsoft CEO Satya Nadella, Alphabet CEO Sundar Pichai, Meta CEO Mark Zuckerberg, Google DeepMind CEO Demis Gassabis, Oracle Board Chairman Larry Ellison, Dell CEO Michael Dell, and Tesla CEO Elon Musk.
NVIDIA did not disclose the cost of the Blackwell cores. Currently, H100s cost between $25,000 and $40,000 per core, and systems based on these cores can cost up to $200,000.
In addition, Nvidia announced partnerships with Apple, Siemens, BYD Co., Johnson & Johnson and introduced the Groot project for humanoid robots.
In particular, Nvidia and Johnson & Johnson will develop new artificial intelligence applications for surgery. It is planned to analyze video recordings of operations and automate document flow, which will allow surgeons to have all the necessary information. Also, artificial intelligence will be used to teach and train surgeons.
To date, H100 cores from NVIDIA are used by almost all companies working on artificial intelligence to train large language models, on the basis of which services such as ChatGPT are created. In less than two years, it has become one of the most valuable companies in the world.
Answers to questions
What is the new core presented by Nvidia?
Why is the Blackwell core considered powerful?
What is the speed of Blackwell nuclei?
What features do Blackwell cores have?
For which companies will the Blackwell core become the basis?
Hashtags
Save a link to this article
Discussion of the topic – The new Nvidia Blackwell chip (B200): the next level of artificial intelligence
The head of Nvidia Jensen Huang presented the new flagship chip Blackwell (B200), based on advanced technologies for artificial intelligence.
Latest comments
9 comments
Write a comment
Your email address will not be published. Required fields are checked *
Emily
I wonder how these new Blackwell chips will change the AI industry. Perhaps we will finally see real leaps in technology!
Felix
Very impressive specs of these new Blackwell chips! It's great that NVIDIA is constantly improving its products.
Sophie
I think these new capabilities from Nvidia will create a new era for data processing and the development of artificial intelligence. Unbelievable!
Matteo
It is always interesting to read about the technological development of the AI industry. NVIDIA is ahead again!
Isabella
I hope that the cost of these new chips will not be too high so that more companies can take advantage of these technologies.
Alexander
I wonder how these new chips from Nvidia will change aspects of how large IT corporations work.
Greta
I was surprised by the speed of the Blackwell chips compared to the H100! NVIDIA promises really impressive achievements.
Hugo
It is important that innovations in AI not only accelerate data processing, but also improve its quality and accuracy.
Lena
Consolidation of the cost of new chips will be an important factor for their widespread use in various fields.