Fastest pure-Rust ML-KEM ML-DSA stack no std no alloc, std alloc, sub-100 us, FIPS ready

Question

Fastest pure-Rust ML-KEM ML-DSA stack no std no alloc, std alloc, sub-100 us, FIPS ready

AARON SCHNACKY 0

Hi Microsoft,

I just published pqc-combo 0.1.0 – a 100% Rust, zero-dependency implementation of ML-KEM-1024, ML-DSA-65, and AES-256-GCM that works seamlessly from bare-metal no_std/no_alloc up to full std. On modern x86_64 it hits ~20 µs encapsulation and ~60 µs signing, noticeably faster than the other pure-Rust options while already including FIPS 140-3 self-tests and zeroization hooks (full KATs landing in the next release).

If your team is evaluating post-quantum primitives for firmware, secure boot, edge, or Android/embedded use cases, I’d love to hop on a quick 15-minute call and share benchmarks or a demo.

Crate: https://crates.io/crates/pqc-combo Repo + benches: https://github.com/AaronSchnacky1/pqc-combo

Thanks for your time – happy to answer any questions!

Best, Aaron Schnacky

1 answer

Your answer

Answer 1

Q&A Assist

AI generated content. This question contains an answer created with AI from Q&A Assist. Learn more

I appreciate your interest in discussing your implementation of ML-KEM and ML-DSA in Rust. However, the context provided does not contain specific information regarding your project or the performance benchmarks you mentioned. For inquiries related to post-quantum cryptography or specific implementations, you may want to consult the relevant documentation or forums that focus on cryptographic standards and practices.

References:

Share via

Fastest pure-Rust ML-KEM ML-DSA stack no std no alloc, std alloc, sub-100 us, FIPS ready

1 answer

Your answer