OpenAI seeks inference chip alternatives amid performance concerns and delayed Nvidia investment talks
OpenAI is pursuing new inference hardware suppliers due to performance concerns. This shift occurs as major investment negotiations with Nvidia face delays.
Insights:
OpenAI has expressed dissatisfaction with certain aspects of the latest artificial intelligence chips from NVIDIA Corporation used for inference and has been actively seeking alternative suppliers since last year. This strategic move centers on the hardware required for inference, the critical stage where a trained model like ChatGPT responds to user inquiries. While NVIDIA Corporationmaintains a dominant position in the market for chips used to train large-scale AI models, the inference segment has emerged as a major competitive front where speed and memory efficiency are paramount.
