How NVIDIA Groq 3 LPX Unlocks Ultrafast Interactivity at Long Context on NVIDIA Vera Rubin
NVIDIA Groq 3 LPX 作为 NVIDIA Vera Rubin 平台的交互式 AI 推理加速器,与核心组件 NVIDIA Vera Rubin NVL72 配合使用。该平台提供高吞吐量和交互性,支持从小模型到大模型、开放及封闭模型的广泛 AI 工作负载。
EVENT DOSSIER
On August 24, 2026, NVIDIA announced that it had expanded its Vera Rubin NVL72 rack-level system to support rapid token generation for agent systems. This upgrade relies on the newly introduced NVIDIA Groq 3 LPX chip, aiming to define the next generation of AI inference through the integration of various levels of collaboration in AI factories, rather than relying on single hardware breakthroughs. As an interactive AI inference accelerator, this platform, used in conjunction with Vera Rubin NVL72, provides high throughput and interactivity, supporting a wide range of AI workloads from small to large models, as well as open and closed models.
NVIDIA announced that it will expand the Vera Rubin NVL72 system to support rapid token generation in agent-based syste…
2 reportsNVIDIA Groq 3 LPX 作为 NVIDIA Vera Rubin 平台的交互式 AI 推理加速器,与核心组件 NVIDIA Vera Rubin NVL72 配合使用。该平台提供高吞吐量和交互性,支持从小模型到大模型、开放及封闭模型的广泛 AI 工作负载。
NVIDIA announced that it will expand the Vera Rubin NVL72 system to support rapid token generation in agent-based systems. This initiative aims to define the next generation of AI inference by integrating the collaborative work at various levels of the AI factory, rather than relying on breakthroughs in single chips, networks, or systems. This update, powered by the NVIDIA Groq 3 LPX chip, marks a further upgrade of the Vera Rubin rack-level system for agent-based systems.