Nvidia slaps $20B Groq tech into massive new LPX racks to speed AI response time
Nvidia announced the integration of Groq's LPU technology into its new Vera Rubin rack systems to accelerate AI inference, specifically the decode phase. This move aims to capture high-value, low-latency inference workloads, with Nvidia projecting potential pricing of up to $45 per million tokens. The announcement is a strategic product roadmap update with no immediate impact on component supply or pricing.
- Related manufacturers
- Nvidia, Groq
- Impact score
- 15 / 100, low impact