ML-KEM-768
ML-KEM (FIPS 203) is a lattice-based key encapsulation mechanism, standardized by NIST in August 2024. It replaces RSA and elliptic-curve Diffie-Hellman for establishing shared secrets in a post-quantum world.
- CPU
- Xeon Platinum 8488C
- Measured on
- c7i.large
- OS / Kernel
- 24.04.4 LTS
- liboqs
- 0.15.0
- Wall clock
- 32m 35s
- CPU steal
- <0.01%
Parameters
- Family
- ML-KEM
- Parameter set
- 768
- Public key
- 1.16 KB
- Private key
- 2.34 KB
- Ciphertext
- 1.06 KB
- Iterations / op
- 1,000
- liboqs key
- ML-KEM-768
Regulatory approval status
Whether this algorithm family is approved/recommended for use under three regimes. Every badge links to its source — click through before citing this elsewhere.
CNSA 2.0 replaces ECDH/RSA key establishment with ML-KEM. Commercial equipment follows CNSA 1.0 until the CNSSP 156-mandated transition (~2025–2030 depending on equipment type).
NSA CNSA 2.0 Cybersecurity Advisory · not directly verified this session, spot-check the sourceListed under §2.4.3 "ML-KEM Key Agreement", one of BSI's recommended quantum-safe asymmetric mechanisms.
BSI TR-02102-1 "Cryptographic Mechanisms: Recommendations and Key Lengths", Version 2026-01 (23 Jan 2026)Recommended at NIST level 5 (preferred) or level 3, using the IND-CCA (active) variant with ephemeral keys where possible — not the parameters modified, and hybridization "strongly emphasized as necessary" in the short/medium term.
ANSSI views on the Post-Quantum Cryptography transition (2023 follow up), §2- Mean
- 11.7 µs
- 95% CI on mean
- ±1.77%
- Median
- 10.6 µs
- p95
- 16.1 µs
- p99
- 24.1 µs
- Stdev
- 3.3 µs
- Min
- 10.2 µs
- Ops/sec
- 85,758
Mean is known to 11.45–11.87 µs across 1,000 iterations. A 95% confidence interval on the mean, not a prediction interval for a single operation. It says how precisely the average is known from this run's samples; the standard deviation says how much the individual samples scattered.
Click any point to view the GitHub Actions run that produced it.
- Mean
- 19.1 µs
- 95% CI on mean
- ±1.85%
- Median
- 18.5 µs
- p95
- 21.2 µs
- p99
- 35.9 µs
- Stdev
- 5.7 µs
- Min
- 13.9 µs
- Ops/sec
- 52,433
Mean is known to 18.72–19.43 µs across 1,000 iterations. A 95% confidence interval on the mean, not a prediction interval for a single operation. It says how precisely the average is known from this run's samples; the standard deviation says how much the individual samples scattered.
Click any point to view the GitHub Actions run that produced it.
- Mean
- 19.8 µs
- 95% CI on mean
- ±1.09%
- Median
- 19.3 µs
- p95
- 22.0 µs
- p99
- 40.4 µs
- Stdev
- 3.5 µs
- Min
- 13.8 µs
- Ops/sec
- 50,603
Mean is known to 19.55–19.98 µs across 1,000 iterations. A 95% confidence interval on the mean, not a prediction interval for a single operation. It says how precisely the average is known from this run's samples; the standard deviation says how much the individual samples scattered.
Click any point to view the GitHub Actions run that produced it.