ConvNext is a Winner!
On Chimera QC Ultra it runs 28x faster than DSP+NPU
ConvNext is one of today’s leading new ML networks. Frankly, it just doesn’t run on most NPUs. So if your AI/ML solution relies on an NPU plus DSP for fallback, you can bet ConvNext needs to run on that DSP as a fallback operation. Our tests of expected performance on a DSP + NPU is dismal – less than one inference per second trying to run the ConvNext backbone if the novel Operators – GeLu and LayerNorm – are not present in the hardwired NPU accelerator and instead those Ops ran on the largest and most powerful DSP core on the market today.
This poor performance makes most of today’s DSP + NPU silicon solutions obsolete. Because that NPU is hardwired, not programmable, new networks cannot efficiently be run.
To read the full article, click here
Related Semiconductor IP
- CXL Controller With HPC Features (Spec Version 3.0 and Above)
- AP Memory VHM Controller
- High-performance, commercial-grade RISC-V CPU Core IP
- SWI3S Verification IP
- Secure Boot Loader
Latest Blogs
- Is Pergrammable a Word?
- Implementing I3C Host Controller support in the open source I3C Core
- SystemLens Gives System Engineers Real-Time Visibility Into Hidden Silicon Issues Under Functional Tests
- As Vehicles Become Data Centers: Why Interface IP Integration Matters in SoC Design
- Implementing a multi-stage secure boot chain on an Efinix Titanium Ti375