Changes between Version 2 and Version 3 of catalina/npu


Ignore:
Timestamp:
10/15/2025 07:54:08 PM (3 days ago)
Author:
Ryan Erbstoesser
Comment:

add links and info

Legend:

Unmodified
Added
Removed
Modified

  • catalina/npu

    v2 v3  
    33The GW9200 and other i.MX95 Catalina SBCs have the following built in NPU for machine learning (ML):
    44 * eIQ® (Edge Intelligence) Neutron N3-1024S NPU (Neural Processing Unit)
    5   * Approximately 2 TOPs but with better benchmarks
     5  * Approximately 2 TOPs
    66  * Designed for high-efficiency and low latency inferences
    77  * Software stack from NXP (Use the NXP Yocto software image)
     8  * Benchmarks can be 2-3x of the i.MX8M Plus NPU, even at the same TOPs
    89 
    910For more serious NPU / TPU calculations, consider a Mini-PCIe or M.2 accelerator that can be added to one of the slots on the GW9200.
    1011
     12Cards:
     13 * https://trac.gateworks.com/wiki/hailoai
     14 * https://trac.gateworks.com/wiki/TPU
     15
     16NXP has a lot of great resources which should be used to learn more about the NPU and how to use it. See links below.
     17
    1118== See Also
    1219 * [wiki:venice/npu Venice NPU Examples (very similar)]
     20 * [https://www.nxp.com/design/design-center/training/TIP-DEEP-DIVE-INTO-IMX-95-NPU-AND-VISION-PIPELINE]
    1321 * [https://www.nxp.com/applications/technologies/ai-and-machine-learning/eiq-neutron-npu]
    1422 * [https://www.nxp.com/design/design-center/software/eiq-ai-development-environment:EIQ]