TLDR:
- Nvidia CEO Jensen Huang shared details about the company’s upcoming Blackwell chip in a recent interview
- The Blackwell chip will be priced between $30,000 and $40,000, similar to the H100
Key Elements of the Article:
Nvidia CEO Jensen Huang recently revealed details about the company’s upcoming Blackwell chip, which was developed at a cost of $10 billion. The new GPU, based on a custom 4NP TSMC process, boasts 208 billion transistors, 192GB of HMB3e memory, and 8TB/s of memory bandwidth. Huang stated that the chip will be priced between $30,000 and $40,000, similar to the H100. However, analysts estimate that the production cost of the Blackwell chip will exceed $6,000, significantly higher than the H100’s estimated $3320 production cost.
Nvidia’s strategy seems to revolve around selling million-dollar AI supercomputers rather than individual chips. The company has introduced various configurations of the Blackwell AI accelerator, including the B100, B200, and GB200 models. These supercomputers, like the NVIDIA GB200 NVL72 rack-scale system and the DGX B200 servers, are expected to incorporate the Blackwell GPUs.
Overall, the Blackwell chip is set to power the next generation of AI, with Nvidia positioning it as a key player in the AI landscape. The technology behind the Blackwell chip represents a significant leap forward in the field of AI and HPC workloads, albeit at a premium price point.