With support for AVX2, AVX-512, and VNNI instruction sets, DeepSparse is validated to work on x86 Intel (Haswell generation and later) and AMD (Zen 2 and later) CPUs running Linux. Mac and Windows require running Linux in a Docker or virtual machine.
Here is a table detailing specific support for some algorithms over different microarchitectures:
| x86 Extension | Microarchitectures | Kernel Sparsity | Sparse Quantization |
|---|---|---|---|
| AMD AVX2 | Zen 2, Zen 3 | optimized | emulated |
| AMD AVX-512 VNNI | Zen 4 | optimized | optimized |
| Intel AVX2 | Haswell, Broadwell, and newer | optimized | emulated |
| Intel AVX-512 | Skylake, Cannon Lake, and newer | optimized | emulated |
| Intel AVX-512 VNNI (DL Boost) | Cascade Lake, Ice Lake, Cooper Lake, Tiger Lake | optimized | optimized |