NVIDIA Groq 3 LPX Now in Full Production With World-Class Speed for Agentic AI
nvidia ·
NVIDIA announced that its Groq 3 LPX interactive AI inference accelerator is now in full production, extending the Vera Rubin platform to speed token generation for agentic AI. The accelerator achieved a record 3,400 output tokens per second in Artificial Analysis benchmarking on the Gemma 4 31B model with a 100,000-token context, the fastest performance ever recorded. Groq 3 LPX reportedly delivers 4x faster responsiveness for agents and latency-sensitive workloads than the nearest alternative platform, enabling tasks like coding to be done in minutes instead of hours. Nebius is the first AI cloud to adopt the platform, planning to offer it through its Nebius Token Factory production inference service, and Groq plans to be among the earliest adopters as well. The product is part of NVIDIA's Vera Rubin AI factory platform, which uses extreme codesign across chips and rack systems to optimize throughput, efficiency, and latency. NVIDIA CEO Jensen Huang said Groq 3 LPX advances the performance frontier for ultrafast token generation and transforms how intelligence is produced.
nvidia