Google published a research blog post on Tuesday about a new compression algorithm for AI models. Within hours, memory stocks were falling. Micron dropped 3 per cent, Western Digital lost 4.7 per cent, and SanDisk fell 5.7 per cent, as investors recalculated how much physical memory the AI industry might actually need.
The algorithm is called TurboQuant, and it addresses one of the most expensive ...
TurboQuant's potential impact on the AI industry is significant, as it addresses a costly bottleneck in running large language models. The algorithm's ability to compress the key-value cache could lead to reduced memory requirements and costs, potentially altering the procurement volumes for AI memory. However, it is uncertain whether this efficiency gain will lead to a decrease in total hardware spending or simply enable more ambitious deployments at roughly the same cost. The history of comput...