FAQGLOBALFIPS-approved cryptographic algorithms

FIPS 203, 204, and 205 post-quantum algorithms What teams need to evidence before making FIPS-approved algorithm claims

FIPS 203 standardizes ML-KEM for key establishment, FIPS 204 standardizes ML-DSA for digital signatures, and FIPS 205 standardizes SLH-DSA for stateless hash-based digital signatures.

This FAQ helps separate algorithm selection from module validation evidence and avoid unsupported claims about post-quantum readiness.

Author
Sorena AI
Published
May 9, 2026
Updated
Jul 25, 2026
Overview

FIPS 203, FIPS 204, and FIPS 205 each standardize a and became effective on August 13, 2024. Treat them as implementation standards for specific algorithms and parameter sets, then separately verify whether the cryptographic module, CAVP algorithm evidence, and CMVP documentation support the claim being made.

Search this module

Find a question or answer quickly

4 of 4 questions
Question 1

What do FIPS 203, FIPS 204, and FIPS 205 cover?

FIPS 203 covers ML-KEM, a module-lattice-based used to establish a shared secret key over a public channel. It is the post-quantum key-establishment standard in this set, not a digital signature standard.

FIPS 204 covers ML-DSA, a module-lattice-based digital signature algorithm for generating and verifying signatures. FIPS 205 covers SLH-DSA, a stateless hash-based digital signature algorithm based on SPHINCS+. Both signature standards apply to public-key-based signature systems operated by federal agencies or operated for them under contract, and their use is also available to private and commercial organizations.

  • Use FIPS 203 when the scoped function is ML-KEM key generation, encapsulation, or decapsulation.
  • Use FIPS 204 when the scoped function is ML-DSA key generation, signature generation, signature verification, or the pre-hash variants supported by the standard.
  • Use FIPS 205 when the scoped function is SLH-DSA key generation, signature generation, signature verification, or approved SHA2/SHAKE parameter-set use.
Question 2

What implementation evidence should teams keep?

The evidence should show the algorithm, parameter set, module boundary, and interface actually used. A bare statement that a product is post-quantum or FIPS-ready is not enough because the NIST standards include parameter-set choices, input checks, approved random bit generation, private-key or decapsulation-key safeguards, and restrictions on exposing testing-only internal interfaces. NIST has posted potential-update errata for FIPS 203 and FIPS 204, so the implementation record should identify the issued standard and the errata version reviewed.

For ML-KEM, record the chosen parameter set and whether encapsulation keys, decapsulation keys, and ciphertexts are checked before use. FIPS 203 maps ML-KEM-512, ML-KEM-768, and ML-KEM-1024 to increasing NIST and recommends ML-KEM-768 as the default unless that is impractical or a different security requirement applies. For ML-DSA, record whether ML-DSA-44, ML-DSA-65, or ML-DSA-87 is used and confirm signature length, public-key length, random seed, and intermediate-data handling requirements. For SLH-DSA, record the SHA2 or SHAKE parameter set, whether the small-signature or fast-signature variant is used, and the key-generation and signing randomness requirements.

  • Map each public claim to a scoped cryptographic function: ML-KEM key establishment, ML-DSA signatures, or SLH-DSA signatures.
  • Store parameter-set evidence, implementation version, module boundary, operating environment, and whether the function is used in an approved mode.
  • Keep test and design evidence separate from certification claims; passing internal tests or using an approved algorithm name does not by itself prove a module is CMVP-validated.
Question 3

How should CAVP and CMVP validation claims be handled?

Use cautious wording. CAVP now publishes validation records for ML-KEM, ML-DSA, and SLH-DSA implementations, and the CMVP implementation guidance adds cryptographic algorithm self-test requirements for these algorithms. Those records support algorithm-conformance claims for the named implementation and tested operational environment. They do not justify saying that a product, library, or cloud service is FIPS 140-3 validated unless the relevant CMVP certificate and Security Policy support that exact module, version, environment, approved mode, and service.

For procurement, customer security reviews, and audit responses, verify the CAVP algorithm entry or CMVP module certificate against the algorithm, parameter set, module name, version, implementation environment, and approved-mode statement. If a supplier only provides roadmap language, test vectors, or an algorithm name, describe it as implementation evidence or migration status rather than validated FIPS evidence.

  • Check the public CAVP validation search for algorithm evidence before citing an algorithm certificate.
  • Check CMVP module documentation before claiming a module is FIPS 140-3 validated or that post-quantum functionality is inside the validated boundary.
  • Record unresolved gaps explicitly, such as missing certificate scope, unsupported parameter set, non-approved mode, unvalidated module boundary, or supplier roadmap-only support.
Question 4

Questions and answers

When should a team use FIPS 203 instead of FIPS 204 or FIPS 205?

Use FIPS 203 when the scoped function is ML-KEM key establishment, including key generation, encapsulation, or decapsulation. Use FIPS 204 when the scoped function is ML-DSA digital signatures, and use FIPS 205 when the scoped function is SLH-DSA stateless hash-based digital signatures. FIPS 203 covers a , while FIPS 204 and FIPS 205 cover digital signature standards.

Which parameter set should teams default to for FIPS 203 and how do the differ?

NIST recommends ML-KEM-768 as the default parameter set because it provides a large security margin at a reasonable performance cost. The three parameter sets are ML-KEM-512 (security category 1), ML-KEM-768 (security category 3), and ML-KEM-1024 (security category 5).

What evidence should support claims about FIPS 203, FIPS 204, or FIPS 205 implementations?

Keep the algorithm, parameter set, module boundary, implementation version, operating environment, and approved mode in scope. CAVP now publishes validation records for ML-KEM, ML-DSA, and SLH-DSA implementations, but a CAVP algorithm record does not by itself prove that a product, library, or module is CMVP-validated. For procurement or audit responses, check the exact CAVP record and, for a module-validation claim, the applicable CMVP certificate and Security Policy.

How should teams handle randomness and side-channel concerns for these post-quantum standards?

Use an approved random bit generator for the random seeds and values that each selected operation requires. ML-KEM key generation and encapsulation require fresh random inputs. ML-DSA and SLH-DSA key generation require randomness, while signing can use deterministic or hedged variants; hedged signing adds fresh randomness and can reduce some side-channel and fault-attack risks. Record which variant the implementation uses and protect or destroy secret intermediate values as the applicable standard requires.

Do the published FIPS 203 and FIPS 204 texts have known errata?

Yes. NIST has posted potential-update errata for FIPS 203 and FIPS 204 and says the identified issues will be corrected in future updates or revisions. Use the issued standards together with the current NIST publication pages and errata when implementing or reviewing conformance; do not silently rewrite the standard or claim that an erratum is already part of a revised FIPS.

Primary sources

References and citations

nvlpubs.nist.gov2 citations
Referenced sections
  • Defines ML-KEM as the FIPS 203 key-encapsulation mechanism and lists ML-KEM-512, ML-KEM-768, and ML-KEM-1024.
  • Supports ML-KEM parameter-set, input-checking, randomness, and default ML-KEM-768 evidence requirements.
"Module-Lattice-Based Key-Encapsulation Mechanism Standard"
More quoted passages
"ML-KEM-768 as the default parameter set"
csrc.nist.gov2 citations
Referenced sections
  • Official publication status and potential-update errata for FIPS 203.
  • Official status page for the issued FIPS 203 text and the errata that NIST says will be addressed in a future update or revision.
nvlpubs.nist.gov2 citations
Referenced sections
  • Defines ML-DSA as the FIPS 204 lattice-based digital signature algorithm and explains its federal signature-system applicability.
  • Supports ML-DSA parameter-set, signature verification, approved random bit generation, and intermediate-data handling evidence.
"Module-Lattice-Based Digital Signature Standard"
More quoted passages
"ML-DSA-44, ML-DSA-65, and ML-DSA-87"
csrc.nist.gov2 citations
Referenced sections
  • Official publication status and potential-update errata for FIPS 204.
  • Official status page for the issued FIPS 204 text and the errata that NIST says will be addressed in a future update or revision.
nvlpubs.nist.gov2 citations
Referenced sections
  • Defines SLH-DSA as the FIPS 205 stateless hash-based digital signature algorithm and identifies its approved parameter-set family.
  • Supports SLH-DSA SHA2/SHAKE parameter-set, key-generation randomness, signing, and verification evidence.
"Stateless Hash-Based Digital Signature Standard"
More quoted passages
"This standard approves 12 parameter sets"
csrc.nist.gov
Referenced sections
  • Public NIST search contains validation records for ML-KEM, ML-DSA, and SLH-DSA implementations and supports checking the named implementation and tested environment.
csrc.nist.gov
Referenced sections
  • Supports the need to account for ML-KEM, ML-DSA, and SLH-DSA cryptographic algorithm self-tests in module validation work.
"Cryptographic Algorithm Self-Test Requirements"
Show all related guides (25)

CAVP Algorithm Validation and ACVP Testing

How CAVP algorithm validation, the ACVP protocol, the ACVTS test system, and CMVP module validation fit together without overstating a FIPS claim.

CAVP Validation Evidence Workflow for FIPS Algorithms

Workflow for collecting CAVP validation evidence and ACVP/ACVTS test records: implementation identity, tested parameters, environments, and CMVP handoff records.

FIPS 180-4 and FIPS 202 secure hash guidance

Choose and evidence SHA-2, SHA-3, and SHAKE use under FIPS 180-4, FIPS 202, CAVP validation, and FIPS 140-3 module claims.

FIPS 186-5 and FIPS 204 digital signatures

Compare FIPS 186-5 classical digital signatures with FIPS 204 ML-DSA, including scope, algorithm choices, key-use limits, and validation evidence boundaries.

FIPS 203 ML-KEM vs RSA and ECDH key establishment

Compare FIPS 203 ML-KEM with RSA and ECDH key-establishment schemes using NIST SP 800-56A, SP 800-56B, CAVP, and CMVP evidence.

FIPS approved algorithm selector workflow

A cited workflow for selecting FIPS and NIST-approved cryptographic algorithms without overstating module validation, CAVP evidence, or approved-mode claims.

FIPS approved mode procurement: certificates, boundaries, and evidence

Procurement guidance for FIPS approved mode claims: how to check CMVP certificates, CAVP evidence, module boundaries, tested environments, and supplier evidence before purchase.

FIPS crypto transition and deprecation tracker

Track FIPS algorithm transitions, withdrawn guidance, CAVP evidence, CMVP module impact, procurement triggers, and approved-mode caveats without overstating validation status.

FIPS cryptographic algorithm selector

Choose between FIPS algorithm standards for AES, SHA-2, SHA-3, digital signatures, ML-KEM, ML-DSA, and SLH-DSA without overstating validation scope.

FIPS KDF and MAC coverage for validated modules

Map FIPS 140-3 KDF and MAC coverage to approved security functions, CAVP evidence, self-tests, service indicators, and module security policy entries.

FIPS Key Management Mapping for Algorithms and SSP Evidence

Map FIPS 140-3 key management requirements to approved algorithms, SSP establishment methods, CAVP evidence, module boundaries, and key-use records.

FIPS Procurement Evidence Review Workflow: CAVP, CMVP, Approved Mode

Review FIPS crypto procurement evidence by separating CAVP algorithm certificates from CMVP module certificates, Security Policy scope, approved mode, operating environment, change impact, and retention records.

FIPS-approved cryptographic algorithms FAQ

Answers to common FIPS algorithm questions: approved security functions, CAVP validation, CMVP module scope, AES modes, SHA-2, SHA-3, signatures, and post-quantum algorithms.

ML-DSA vs ECDSA under FIPS 204 and FIPS 186-5

Compare ML-DSA and ECDSA for FIPS-aligned digital signature designs, including parameter choices, key handling, CAVP algorithm evidence, and CMVP module boundaries.

Post-quantum FIPS 203, 204, and 205: ML-KEM, ML-DSA, and SLH-DSA

An official source guide to the three NIST post-quantum FIPS standards: when ML-KEM, ML-DSA, and SLH-DSA apply, what evidence to keep, and how CAVP and CMVP claims differ.

Post-Quantum Migration for FIPS Cryptography

Plan post-quantum migration for FIPS cryptography by separating ML-KEM key establishment, ML-DSA and SLH-DSA signatures, CAVP algorithm evidence, and CMVP module validation boundaries.

Post-Quantum Migration Tracker for FIPS 203, 204, and 205

Track post-quantum cryptography migration evidence for FIPS 203 ML-KEM, FIPS 204 ML-DSA, FIPS 205 SLH-DSA, CAVP algorithm certificates, and CMVP module boundaries.

SHA-2 vs SHA-3 under FIPS 180-4 and FIPS 202

Compare SHA-2 and SHA-3 for FIPS use: approved functions, validation evidence, compatibility, procurement checks, and when migration is not required.

TLS use-case mapping for FIPS algorithm evidence

Map TLS uses to FIPS algorithm, CAVP, CMVP, approved-mode, certificate-authority, and evidence checks without overstating protocol validation claims.