#1
#1 out of 13.37%
technology8h ago
Nvidia may soon unveil a brand-new AI chip. A closer look at the $20 billion bet to make it happen
- Nvidia licenses Groq tech and hires key Groq leadership in a $20 billion move to boost inference hardware.
- Nvidia aims to use Groq’s LPUs to create a new, faster inference-focused chip for AI workloads.
- The deal could expand Nvidia’s platform beyond GPUs, mirroring the Mellanox integration for networking.
- Groq’s founder Jonathan Ross joins Nvidia as chief software architect after moving from Groq.
- Analysts anticipate Nvidia leveraging Groq to improve inference performance and cost efficiency.
- Nvidia’s strategy emphasizes platform agnosticism, aiming to work with various architectures and accelerators.
- Nvidia’s broader AI push includes expansions into standalone CPU designs like Grace and Vera Rubin.
- GTC conference events will feature updates on Nvidia’s Vera Rubin and CPU roadmap.
- Nvidia’s data center revenue remains a key driver, with substantial growth reported in recent quarters.
- The Groq licensing move signals continued consolidation and collaboration in AI hardware.
- Industry context shows rising demand for inference chips as AI adoption expands across data centers.
Vote 0