Kyber

Algorithm type: Key encapsulation mechanism.
Main cryptographic assumption: Module LWE+R with base ring Z[x]/(3329, x^256+1).
Principal submitters: Peter Schwabe.
Auxiliary submitters: Roberto Avanzi, Joppe Bos, Léo Ducas, Eike Kiltz, Tancrède Lepoint, Vadim Lyubashevsky, John M. Schanck, Gregor Seiler, Damien Stehlé.
Authors’ website: https://pq-crystals.org/
Specification version: NIST Round 3 submission.
Primary Source:
- Source: https://github.com/pq-crystals/kyber/commit/441c0519a07e8b86c8d079954a6b10bd31d29efc with copy_from_upstream patches
- Implementation license (SPDX-Identifier): CC0-1.0 or Apache-2.0
Optimized Implementation sources: https://github.com/pq-crystals/kyber/commit/441c0519a07e8b86c8d079954a6b10bd31d29efc with copy_from_upstream patches
- oldpqclean-aarch64:
  - Source: https://github.com/PQClean/PQClean/commit/8e220a87308154d48fdfac40abbb191ac7fce06a with copy_from_upstream patches
  - Implementation license (SPDX-Identifier): CC0-1.0 and (CC0-1.0 or Apache-2.0) and (CC0-1.0 or MIT) and MIT
Formally-verified Implementation sources:
- libjade:
  - Source: https://github.com/formosa-crypto/libjade/tree/release/2023.05-2 with copy_from_upstream patches
  - Implementation license (SPDX-Identifier): CC0-1.0 OR Apache-2.0

Parameter set summary

Parameter set	Parameter set alias	Security model	Claimed NIST Level	Public key size (bytes)	Secret key size (bytes)	Ciphertext size (bytes)	Shared secret size (bytes)
Kyber512	NA	IND-CCA2	1	800	1632	768	32
Kyber768	NA	IND-CCA2	3	1184	2400	1088	32
Kyber1024	NA	IND-CCA2	5	1568	3168	1568	32

Implementation source	Identifier in upstream	Supported architecture(s)	Supported operating system(s)	CPU extension(s) used	No branching-on-secrets claimed?	No branching-on-secrets checked by valgrind?	Large stack usage?‡
Primary Source	ref	All	All	None	True	True	False
Primary Source	avx2	x86_64	Linux,Darwin	AVX2,BMI2,POPCNT	True	True	False
oldpqclean-aarch64	aarch64	ARM64_V8	Linux,Darwin	None	True	False	False
libjade	ref	x86_64	Linux,Darwin	None	True	False	False
libjade	avx2	x86_64	Linux,Darwin	AVX2,BMI2,POPCNT	True	False	False

Are implementations chosen based on runtime CPU feature detection? Yes.

‡For an explanation of what this denotes, consult the Explanation of Terms section at the end of this file.

Implementation source	Identifier in upstream	Supported architecture(s)	Supported operating system(s)	CPU extension(s) used	No branching-on-secrets claimed?	No branching-on-secrets checked by valgrind?	Large stack usage?
Primary Source	ref	All	All	None	True	True	False
Primary Source	avx2	x86_64	Linux,Darwin	AVX2,BMI2,POPCNT	True	True	False
oldpqclean-aarch64	aarch64	ARM64_V8	Linux,Darwin	None	True	False	False
libjade	ref	x86_64	Linux,Darwin	None	True	False	False
libjade	avx2	x86_64	Linux,Darwin	AVX2,BMI2,POPCNT	True	False	False

Are implementations chosen based on runtime CPU feature detection? Yes.

Implementation source	Identifier in upstream	Supported architecture(s)	Supported operating system(s)	CPU extension(s) used	No branching-on-secrets claimed?	No branching-on-secrets checked by valgrind?	Large stack usage?
Primary Source	ref	All	All	None	True	True	False
Primary Source	avx2	x86_64	Linux,Darwin	AVX2,BMI2,POPCNT	True	True	False
oldpqclean-aarch64	aarch64	ARM64_V8	Linux,Darwin	None	True	False	False

Are implementations chosen based on runtime CPU feature detection? Yes.

Large Stack Usage: Implementations identified as having such may cause failures when running in threads or in constrained environments.