COMPUTE & AI

Edge AI

Running neural-network inference on local hardware such as NPUs, GPUs and accelerator modules, usually quoted in TOPS for a given numeric precision.

  • Matching partsIn the catalog
  • 2 categoriesRelated families
  • 8 solutionsWhere it is used

What to check when selecting hardware

  • TOPS and the precision it refers to (for example INT8)
  • Host interface (M.2, PCIe, integrated) and power budget
  • Supported frameworks, model-conversion toolchain and SDK

Parts in the catalog

Parts are matched to this technology by category or description. Confirm support in the datasheet.

See all parts