‘AI 추론 가속기’ 키워드 기사 2 articles
NVIDIA Begins Mass Production of Groq 3 LPX…Achieves 3,400 Tokens Per Second Inference Performance
NVIDIA has started mass production of the Groq 3 LPX AI inference accelerator. Based on the Vera Rubin NVL72 platform, this product achieves 3,400 out…
2026.08.25 by 배종인 기자
Puriosa AI to Supply RNGD to Sweden's 15MW AI Data Center
Domestic AI semiconductor company Puriosa AI will supply its AI inference accelerator RNGD to a 15MW-scale AI data center in Stockholm, Sweden. In col…
2026.08.05 by 배종인 기자