Tenstorrent Grayskull
Realizes: AI training acceleration
Tenstorrent Grayskull is a tile-based architecture of compute tiles with systolic arrays paired with on-tile high-bandwidth memory to deliver massive data-parallel tensor math and training throughput for large neural networks.
Examples
🔗
Tenstorrent Grayskull board
Grayskull board deploys the tile-based ASIC with on-package high-bandwidth memory to tackle large-scale transformer and generative model training.
Mixed-precision transformer training, sparse-dense tensor contractions
high
datacenter
unknown