Nvidia Bets on Real-Time AI With Inference-Focused Chip
Nvidia is prepping a new processor optimized for AI inference to speed real-time responses, to be unveiled at its GTC conference; the chip targets inference workloads rather than training, aims to expand beyond Nvidia’s training chips as OpenAI and other users seek faster, energy-efficient AI services amid rising competition from Alphabet and Amazon; TipRanks cites a price target implying substantial upside for NVDA.