Lightning-fast Kyber (ML-KEM) implementations as specified in FIPS-203 are now here as of wolfSSL release 5.7.4:

ARM32 v4 to v8 base assembly instructions for Kyber (ML-KEM); PR 8040

Even faster NEON instructions are under development!

Aarch64 implementations of Kyber (ML-KEM) functions; PR 7998 SHA-3 assembly implementations used by Kyber (ML-KEM); PR 7998 ARMv7E-M/ARMv7-M assembly instructions for Kyber (ML-KEM); PR 7706

If you’re worried about the performance and speed of post-quantum Kyber (ML-KEM); you shouldn’t be! Even without these lightning fast optimizations, ML-DSA beats ECDH. You can see our benchmarks comparing ECDH against ML-DSA on ARM Cortex M4 here. Now, we give it that extra bit of oomph that leaves our competition in the dust! Come on out and try it for yourself! If you have questions about any of the above, please contact us at facts@wolfSSL.com or +1 425 245 8247. Download wolfSSL Now