Skip to main content

NIST Post-Quantum Cryptography Standards: Complete Timeline and Guide

~7 min readBeginner

NIST spent eight years evaluating post-quantum cryptography candidates — the longest cryptographic standardization process in its history. In August 2024, the results became final: three new federal standards that will replace RSA, ECDSA, and ECDH across every industry that depends on public-key cryptography.

This article covers what was standardized, why it matters, and what the compliance timeline looks like for your organization.

The Standardization Process

In 2016, NIST issued a public call for post-quantum cryptographic algorithms. The response was enormous: 82 submissions from 25 countries, representing the global cryptographic research community. Over the next eight years, NIST conducted the most rigorous public evaluation of cryptographic algorithms ever attempted.

NIST Post-Quantum Cryptography Standardization

2016Call for Proposals

NIST issues a public call for post-quantum cryptographic algorithms. 82 submissions received from 25 countries.

2017Round 1 — Initial Review

69 candidates accepted for evaluation after initial screening. Public review and cryptanalysis begins.

2019Round 2 — Narrowed Field

26 candidates advance. Community cryptanalysis eliminates algorithms with discovered weaknesses.

2020Round 3 — Finalists Selected

15 finalists and alternates announced. Kyber, Dilithium, and SPHINCS+ emerge as frontrunners.

2022Winners Announced

NIST selects Kyber (KEM), Dilithium (signatures), and SPHINCS+ (backup signatures) for standardization.

Aug 2024Final Standards Published

FIPS 203 (ML-KEM), FIPS 204 (ML-DSA), and FIPS 205 (SLH-DSA) published as official federal standards.

2025+Industry Adoption

Migration begins across government, finance, healthcare, and technology sectors. CNSA 2.0 mandates take effect.

Why does NIST's decision matter globally? Because NIST standards become the foundation for interoperability. When your bank, your cloud provider, your TLS library, and your hardware security module all need to agree on which algorithms to support, they follow NIST. The European Union (ETSI), Japan (CRYPTREC), and other national standards bodies have aligned their recommendations with NIST's selections.

The process was deliberately public and adversarial. Every candidate was published openly. Researchers worldwide attempted to break each algorithm. Candidates that fell to cryptanalysis were eliminated. The survivors endured eight years of the world's best cryptographic minds trying to find weaknesses.

The Three Standards

FIPS 203: ML-KEM (Based on Kyber)

What it does: Key Encapsulation Mechanism (KEM) — securely establishes a shared secret between two parties.

Use case: Replaces RSA and ECDH for key exchange. When two systems need to agree on a symmetric encryption key, ML-KEM provides a quantum-safe way to do it.

Parameter sets:

  • ML-KEM-512 — NIST Level 1 (roughly AES-128 equivalent). Smallest keys, fastest performance.
  • ML-KEM-768 — NIST Level 3 (roughly AES-192 equivalent). The most widely recommended parameter set, balancing security and performance.
  • ML-KEM-1024 — NIST Level 5 (roughly AES-256 equivalent). Maximum security, larger keys.

The underlying mathematics is the Module-LWE (Module Learning With Errors) problem — a lattice-based hard problem that no known quantum algorithm can efficiently solve. The Kyber (ML-KEM) deep dive covers the intuition behind lattice cryptography.

FIPS 204: ML-DSA (Based on Dilithium)

What it does: Digital signatures — proves that a message was signed by a specific key holder and has not been tampered with.

Use case: Replaces RSA and ECDSA signatures for code signing, document signing, TLS certificate verification, and API response integrity.

Parameter sets:

  • ML-DSA-44 — NIST Level 2. Smallest signatures.
  • ML-DSA-65 — NIST Level 3. The most widely recommended parameter set.
  • ML-DSA-87 — NIST Level 5. Maximum security.

ML-DSA is also based on the Module-LWE problem family, sharing its mathematical foundation with ML-KEM. The Dilithium (ML-DSA) deep dive explains how lattice-based signatures work.

FIPS 205: SLH-DSA (Based on SPHINCS+)

What it does: Stateless hash-based digital signatures — an alternative signature scheme based on different mathematical foundations.

Use case: Conservative backup option for signatures. If a future breakthrough weakens lattice-based cryptography, SLH-DSA provides an independent fallback based solely on hash functions.

Key tradeoff: SLH-DSA signatures are significantly larger and slower than ML-DSA. It is intended as insurance, not as a primary replacement for ECDSA. Most organizations will use ML-DSA for day-to-day operations and may keep SLH-DSA available as a contingency.

PropertyKyber768Dilithium3
NIST NameML-KEM-768ML-DSA-65
NIST StandardFIPS 203FIPS 204
OperationKey Encapsulation (Encrypt/Decrypt)Digital Signatures (Sign/Verify)
Security LevelNIST Level 3 (~AES-192)NIST Level 3 (~AES-192)
Public Key Size1,184 bytes1,952 bytes
Private Key Size2,400 bytes4,000 bytes
Ciphertext Size1,088 bytes3,293 bytes
Shared Secret Size32 bytesN/A
Latency Target< 15ms (p95)< 30ms (p95)

What Qpher Implements

Qpher implements the Level 3 parameter sets of both primary standards:

  • Kyber768 (ML-KEM-768) for key encapsulation — quantum-safe encryption
  • Dilithium3 (ML-DSA-65) for digital signatures — quantum-safe integrity

Why Level 3? It represents the sweet spot between security and API performance. Level 3 provides security roughly equivalent to AES-192 against both classical and quantum attacks. This exceeds the security requirements of all but the most sensitive government applications, while maintaining the low latency that API consumers expect: KEM operations under 15ms (p95) and signature operations under 30ms (p95).

For a detailed look at how Qpher implements these standards, see the Core Concepts page.

Compliance Timeline

The publication of FIPS 203, 204, and 205 starts a compliance clock. Government mandates drive the initial wave, but private-sector requirements follow closely behind.

YearMilestone
Aug 2024NIST publishes FIPS 203, 204, 205 — standards are final
2025NSA CNSA 2.0 requires PQC for all new national security systems
2026Federal agencies begin active migration of existing systems
2030CNSA 2.0 target: PQC-only for software and firmware signing
2033CNSA 2.0 target: PQC-only for web browsers, cloud services, and networking
2035Full PQC migration expected across regulated industries
Supply chain pressure

You do not have to be a government contractor to be affected. When banks, healthcare systems, and cloud providers adopt PQC mandates, they require it from their vendors and partners. If your software handles sensitive data for any regulated customer, PQC compliance will eventually flow down to you through contractual requirements.

The compliance timeline is not a suggestion. Organizations subject to CNSA 2.0 (U.S. national security systems) face hard deadlines. Federal civilian agencies follow OMB mandates with their own timelines. Private-sector industries — finance (PCI DSS), healthcare (HIPAA), and others — will issue their own PQC requirements as standards mature.

The organizations that begin migration planning now will have a smooth, multi-year transition. Those that wait until mandates take effect will face rushed, expensive, and error-prone migrations under regulatory pressure.

Next Steps

The beginner learning path is complete. You now understand:

  1. How quantum computers threaten encryption
  2. Why the threat is urgent today (HNDL)
  3. What NIST standardized and when you need to comply (this article)

Ready to understand how the algorithms work under the hood? The intermediate path starts with a developer-friendly deep dive into Kyber768 — how lattice cryptography makes quantum-safe encryption possible.

For a hands-on introduction to Qpher's PQC API, see the Core Concepts overview.