Qualcomm made an AI inferencing card with 128GB RAM without using HBM: https://w...

		ryao 8 months ago \| parent \| context \| favorite \| on: Intel announces Arc B-series "Battlemage" discrete... Qualcomm made an AI inferencing card with 128GB RAM without using HBM: https://www.qualcomm.com/news/onq/2023/11/introducing-qualco... Would someone with “basic ‘knowledge’ of hardware” explain why a GPU cannot be made with the same memory configuration?