NVIDIA Groq 3 LPX in full production, hits 3,400 tokens/sec

NVIDIA announced Groq 3 LPX, an interactive inference accelerator for Vera Rubin, is in full production. In an Artificial Analysis benchmark on Gemma 4 31B, it delivered 3,400 output tokens per second at 100K context, 4x faster than the nearest alternative. Nebius is the first AI cloud to adopt it.