noble-curves

Audited & minimal JS implementation of elliptic curve cryptography.

• 🔒 Audited by independent security firms
• 🔻 Tree-shakeable: unused code is excluded from your builds
• 🏎 Fast: hand-optimized for caveats of JS engines
• 🔍 Reliable: cross-library / wycheproof tests and fuzzing ensure correctness
• ➰ Weierstrass, Edwards, Montgomery curves; ECDSA, EdDSA, Schnorr, BLS signatures
• ✍️ ECDH, hash-to-curve, OPRF, FROST, Poseidon hash, FFT
• 🔖 Non-repudiation (SUF-CMA, SBS) & consensus-friendliness (ZIP215) in ed25519, ed448
• 🥈 Optional, friendly wrapper over native WebCrypto
• 🪶 32KB (gzipped) including bundled hashes, 11KB for single-curve build

Curves have 5kb sister projects secp256k1 & ed25519. They have smaller attack surface, but less features.

Take a glance at GitHub Discussions for questions and support.

This library belongs to noble cryptography

noble cryptography — high-security, easily auditable set of contained cryptographic libraries and tools.

• Zero or minimal dependencies
• Highly readable TypeScript / JS code
• PGP-signed releases and transparent NPM builds
• All libraries: ciphers, curves, hashes, post-quantum, 5kb secp256k1 / ed25519
• Check out the homepage for reading resources, documentation, and apps built with noble

Usage

npm install @noble/curves

deno add jsr:@noble/curves

We support all major platforms and runtimes. For React Native, you may need a polyfill for getRandomValues. A standalone file noble-curves.js is also available.

// import * from '@noble/curves'; // Error: use sub-imports, to ensure small app size
import { secp256k1, schnorr } from '@noble/curves/secp256k1.js';
import { ed25519, ed25519ph, ed25519ctx, x25519, ristretto255 } from '@noble/curves/ed25519.js';
import { ed448, ed448ph, x448, decaf448 } from '@noble/curves/ed448.js';
import { p256, p384, p521 } from '@noble/curves/nist.js';
import { bls12_381 } from '@noble/curves/bls12-381.js';
import { bn254 } from '@noble/curves/bn254.js';
import { jubjub, babyjubjub, brainpoolP256r1, brainpoolP384r1, brainpoolP512r1 } from '@noble/curves/misc.js';

// hash-to-curve
import { secp256k1_hasher } from '@noble/curves/secp256k1.js';
import { p256_hasher, p384_hasher, p521_hasher } from '@noble/curves/nist.js';
import { ristretto255_hasher } from '@noble/curves/ed25519.js';
import { decaf448_hasher } from '@noble/curves/ed448.js';

// OPRFs
import { p256_oprf, p384_oprf, p521_oprf } from '@noble/curves/nist.js';
import { ristretto255_oprf } from '@noble/curves/ed25519.js';
import { decaf448_oprf } from '@noble/curves/ed448.js';

// utils
import { bytesToHex, hexToBytes, concatBytes } from '@noble/curves/utils.js';
import { Field } from '@noble/curves/abstract/modular.js';
import { weierstrass, ecdsa } from '@noble/curves/abstract/weierstrass.js';
import { edwards, eddsa } from '@noble/curves/abstract/edwards.js';
import { poseidon, poseidonSponge } from '@noble/curves/abstract/poseidon.js';
import { FFT, poly } from '@noble/curves/abstract/fft.js';

• Examples
  • ECDSA, EdDSA, Schnorr signatures
    • secp256k1, p256, p384, p521, ed25519, ed448, brainpool
    • ristretto255, decaf448
    • Prehashed signing
    • Recovering public keys from signatures
    • Hedged ECDSA with noise
    • Consensus-friendliness vs e-voting
  • ECDH: Diffie-Hellman shared secrets
  • webcrypto: Friendly wrapper
  • BLS signatures, bls12-381, bn254 aka alt_bn128
  • Hashing to curve points
  • OPRFs
  • FROST threshold signatures
  • Poseidon hash
  • Fast Fourier Transform
  • utils
• Internals
  • Elliptic curve Point math
  • modular: Modular arithmetics & finite fields
  • weierstrass: Custom Weierstrass curve
  • edwards: Custom Edwards curve
  • Custom ECDSA instance
• Security
• Speed
• Contributing & testing
• Upgrading

ECDSA, EdDSA, Schnorr signatures

secp256k1, p256, p384, p521, ed25519, ed448, brainpool

import { secp256k1, schnorr } from '@noble/curves/secp256k1.js';
import { p256, p384, p521 } from '@noble/curves/nist.js';
import { ed25519 } from '@noble/curves/ed25519.js';
import { ed448 } from '@noble/curves/ed448.js';
import { brainpoolP256r1, brainpoolP384r1, brainpoolP512r1 } from '@noble/curves/misc.js';
for (const curve of [
  secp256k1, schnorr,
  p256, p384, p521,
  ed25519, ed448,
  brainpoolP256r1, brainpoolP384r1, brainpoolP512r1
]) {
  const { secretKey, publicKey } = curve.keygen();
  const msg = new TextEncoder().encode('hello noble');
  const sig = curve.sign(msg, secretKey);
  const isValid = curve.verify(sig, msg, publicKey);
  console.log(curve, secretKey, publicKey, sig, isValid);
}

// Specific private key
import { hexToBytes } from '@noble/curves/utils.js';
const secret2 = hexToBytes('46c930bc7bb4db7f55da20798697421b98c4175a52c630294d75a84b9c126236');
const pub2 = secp256k1.getPublicKey(secret2);

ECDSA signatures use deterministic k, conforming to RFC 6979. EdDSA conforms to RFC 8032. Schnorr (secp256k1-only) conforms to BIP 340.

Messages are always hashed first.

ristretto255, decaf448

import { ristretto255, ristretto255_hasher, ristretto255_oprf } from '@noble/curves/ed25519.js';
import { decaf448, decaf448_hasher, decaf448_oprf } from '@noble/curves/ed448.js';

console.log(ristretto255.Point, decaf448.Point);

Check out RFC 9496 more info on ristretto255 & decaf448. Check out separate documentation for Point, hasher and oprf.

Prehashed signing

import { secp256k1 } from '@noble/curves/secp256k1.js';
import { keccak_256 } from '@noble/hashes/sha3.js';
const { secretKey } = secp256k1.keygen();
const msg = new TextEncoder().encode('hello noble');
// prehash: true (default) - hash using secp256k1.hash (sha256)
const sig = secp256k1.sign(msg, secretKey);
// prehash: false - hash using custom hash
const sigKeccak = secp256k1.sign(keccak_256(msg), secretKey, { prehash: false });

Default sign() and verify() behavior (prehash: true) applies built-in hash function to message first. For secp256k1 that's sha256, for p521 that's sha512.

Providing prehash: false allows user to specify their own hash function (e.g. use secp256k1 + keccak_256).

[!NOTE] Previously, in noble-curves v1, prehash: false was the default. Some other libraries (like libsecp256k1) have no prehashing.

Recovering public keys from signatures

import { secp256k1 } from '@noble/curves/secp256k1.js';
const { secretKey, publicKey } = secp256k1.keygen();
const msg = new TextEncoder().encode('hello noble');
const sigRec = secp256k1.sign(msg, secretKey, { format: 'recovered' });
const publicKey_ = secp256k1.recoverPublicKey(sigRec, msg); // == publicKey

// recovered sig is compact sig with an extra byte
const sigNoRec = secp256k1.sign(msg, secretKey, { format: 'compact' });
// sigNoRec == sigRec.slice(1)

// Signature instance
const sigInstance = secp256k1.Signature.fromBytes(sigRec, 'recovered');

Public key recovery - only supported with ECDSA.

[!NOTE] Key recovery is a simple math operation. There are no guarantees the signing was actually done. It's possible to forge signature and msg hash (r, s, h), which would recover into a random public key, but it's not feasible to find m which would lead to this specific forged h.

Hedged ECDSA with noise

import { secp256k1 } from '@noble/curves/secp256k1.js';
const { secretKey } = secp256k1.keygen();
const msg = new TextEncoder().encode('hello noble');
// extraEntropy: false - default, hedging disabled
const sigNoisy = secp256k1.sign(msg, secretKey);
// extraEntropy: true - fetch 32 random bytes from CSPRNG
const sigNoisyA = secp256k1.sign(msg, secretKey, { extraEntropy: true });
// extraEntropy: bytes - specific extra entropy
const ent = Uint8Array.from([0xca, 0xfe, 0x01, 0x23]);
const sigNoisy2 = secp256k1.sign(msg, secretKey, { extraEntropy: ent });

ECDSA sign() allows providing extraEntropy, which switches sig generation to hedged mode.

By default, ECDSA signatures are generated deterministically, following RFC 6979. However, purely deterministic signatures are vulnerable to fault attacks. Newer signature schemes, such as BIP340 schnorr, switched to hedged signatures because of this. Hedging is basically incorporating some randomness into sig generation process.

For more info, check out Deterministic signatures are not your friends, RFC 6979 section 3.6, and cfrg-det-sigs-with-noise draft.

Consensus-friendliness vs e-voting

import { ed25519 } from '@noble/curves/ed25519.js';
const { secretKey, publicKey } = ed25519.keygen();
const msg = new TextEncoder().encode('hello noble');
const sig = ed25519.sign(msg, secretKey);
// zip215: true
const isValid = ed25519.verify(sig, msg, publicKey);
// SBS / e-voting / RFC8032 / FIPS 186-5
const isValidRfc = ed25519.verify(sig, msg, publicKey, { zip215: false });

[!NOTE] Most other libraries don't have SUF-CMA & SBS - less optimal choice for their security.

In ed25519, there is an ability to choose between consensus-friendliness vs e-voting mode.

• zip215: true (default) uses the more permissive, consensus-friendly verification rules defined in ZIP215.
• zip215: false enforces strict RFC 8032 / FIPS 186-5 verification and adds SBS-based non-repudiation, which is useful for:
  • Contract signing: prevents a signer from later claiming they signed a different document
  • E-voting: stops voters from choosing keys that let them repudiate their vote
  • Blockchains: avoids signatures valid for multiple transactions (e.g., amount X also validating amount Y)

Both modes have SUF-CMA (strong unforgeability under chosen message attacks). See Taming the many EdDSAs for more info.

ECDH: Diffie-Hellman shared secrets

import { secp256k1 } from '@noble/curves/secp256k1.js';
import { x25519 } from '@noble/curves/ed25519.js';
import { x448 } from '@noble/curves/ed448.js';
import { p256, p384, p521 } from '@noble/curves/nist.js';

for (const curve of [secp256k1, x25519, x448, p256, p384, p521]) {
  const alice = curve.keygen();
  const bob = curve.keygen();
  const sharedKey = curve.getSharedSecret(alice.secretKey, bob.publicKey);
  console.log('alice', alice, 'bob', bob, 'shared', sharedKey);
}

// x25519 & x448 specific methods
import { ed25519 } from '@noble/curves/ed25519.js';
const alice = ed25519.keygen();
const bob = ed25519.keygen();
const aliceSecX = ed25519.utils.toMontgomerySecret(alice.secretKey);
const bobPubX = ed25519.utils.toMontgomery(bob.publicKey);
const sharedKey = x25519.getSharedSecret(aliceSecX, bobPubX);

We provide ECDH over all Weierstrass curves, and over 2 Montgomery curves X25519 (Curve25519) & X448 (Curve448), conforming to RFC 7748.

In Weierstrass curves, shared secrets:

• Include y-parity bytes: use key.slice(1) to strip it
• Are not hashed: use hashing or KDF on top, like sha256(shared) or hkdf(shared)

webcrypto: Friendly wrapper

[!NOTE] Webcrypto methods are always async.

webcrypto signatures

import { ed25519, ed448, p256, p384, p521 } from '@noble/curves/webcrypto.js';

(async () => {
  for (let [name, curve] of Object.entries({ p256, p384, p521, ed25519, ed448 })) {
    console.log('curve', name);
    if (!await curve.isSupported()) {
      console.log('is not supported, skipping');
      continue;
    }
    const keys = await curve.keygen();
    const msg = new TextEncoder().encode('hello noble');
    const sig = await curve.sign(msg, keys.secretKey);
    const isValid = await curve.verify(sig, msg, keys.publicKey);
    console.log({
      keys, msg, sig, isValid
    });
  }
})();

webcrypto ecdh

import { p256, p384, p521, x25519, x448 } from '@noble/curves/webcrypto.js';

(async () => {
  for (let [name, curve] of Object.entries({ p256, p384, p521, x25519, x448 })) {
    console.log('curve', name);
    if (!await curve.isSupported()) {
      console.log('is not supported, skipping');
      continue;
    }
    const alice = await curve.keygen();
    const bob = await curve.keygen();
    const shared = await curve.getSharedSecret(alice.secretKey, bob.publicKey);
    const shared2 = await curve.getSharedSecret(bob.secretKey, alice.publicKey);
    console.log({shared});
  }
})();

Key conversion from noble to webcrypto and back

import { p256 as p256n } from '@noble/curves/nist.js';
import { p256 } from '@noble/curves/webcrypto.js';
(async () => {
  const nobleKeys = p256n.keygen();
  // convert noble keys to webcrypto
  const webKeys = {
    secretKey: await p256.utils.convertSecretKey(nobleKeys.secretKey, 'raw', 'pkcs8'),
    publicKey: await p256.utils.convertPublicKey(nobleKeys.publicKey, 'raw', 'spki')
  };
  // convert webcrypto keys to noble
  const nobleKeys2 = {
    secretKey: await p256.utils.convertSecretKey(webKeys.secretKey, 'pkcs8', 'raw'),
    publicKey: await p256.utils.convertPublicKey(webKeys.publicKey, 'spki', 'raw')
  };
})();

Check out micro-key-producer for pure JS key conversion utils.

BLS signatures, bls12-381, bn254 aka alt_bn128

import { bls12_381 } from '@noble/curves/bls12-381.js';

// G1 pubkeys, G2 sigs
const blsl = bls12_381.longSignatures;
const { secretKey, publicKey } = blsl.keygen();
// const publicKey = blsl.getPublicKey(secretKey);
const msg = new TextEncoder().encode('hello noble');
// default DST
const msgp = blsl.hash(msg);
// custom DST (Ethereum)
const msgpd = blsl.hash(msg, 'BLS_SIG_BLS12381G2_XMD:SHA-256_SSWU_RO_POP_');
const signature = blsl.sign(msgp, secretKey);
const isValid = blsl.verify(signature, msgp, publicKey);
console.log('long', { publicKey, signature, isValid });

// G1 sigs, G2 pubkeys
const blss = bls12_381.shortSignatures;
const publicKey2 = blss.getPublicKey(secretKey);
const msgp2 = blss.hash(msg, 'BLS_SIG_BLS12381G1_XMD:SHA-256_SSWU_RO_NUL_');
const signature2 = blss.sign(msgp2, secretKey);
const isValid2 = blss.verify(signature2, msgp2, publicKey2);
console.log({ publicKey2, signature2, isValid2 });

// Aggregation
const aggregatedKey = bls12_381.longSignatures.aggregatePublicKeys([
  blsl.getPublicKey(bls12_381.utils.randomSecretKey()),
  blsl.getPublicKey(bls12_381.utils.randomSecretKey()),
]);
// const aggregatedSig = bls.aggregateSignatures(sigs)

// Pairings, with and without final exponentiation
// bls.pairing(PointG1, PointG2);
// bls.pairing(PointG1, PointG2, false);
// bls.fields.Fp12.finalExponentiate(bls.fields.Fp12.mul(PointG1, PointG2));

// Others
// bls.G1.Point.BASE, bls.G2.Point.BASE;
// bls.fields.Fp, bls.fields.Fp2, bls.fields.Fp12, bls.fields.Fr;

See abstract/bls. For example usage, check out the implementation of BLS EVM precompiles.

The BN254 API mirrors BLS. The curve was previously called alt_bn128. The implementation is compatible with EIP-196 and EIP-197.

For BN254 usage, check out the implementation of bn254 EVM precompiles. We don't implement Point methods toBytes. To work around this limitation, has to initialize points on their own from BigInts. Reason it's not implemented is because there is no standard. Points of divergence:

• Endianness: LE vs BE (byte-swapped)
• Flags as first hex bits (similar to BLS) vs no-flags
• Imaginary part last in G2 vs first (c0, c1 vs c1, c0)

hash-to-curve: hashing to curve points

import { bls12_381 } from '@noble/curves/bls12-381.js';
import { ed25519_hasher, ristretto255_hasher } from '@noble/curves/ed25519.js';
import { decaf448_hasher, ed448_hasher } from '@noble/curves/ed448.js';
import { p256_hasher, p384_hasher, p521_hasher } from '@noble/curves/nist.js';
import { secp256k1_hasher } from '@noble/curves/secp256k1.js';

const h = {
  secp256k1_hasher,
  p256_hasher, p384_hasher, p521_hasher,
  ed25519_hasher,
  ed448_hasher,
  ristretto255_hasher,
  decaf448_hasher,
  bls_G1: bls12_381.G1,
  bls_G2: bls12_381.G2
};

const msg = Uint8Array.from([0xca, 0xfe, 0x01, 0x23]);
console.log('msg', msg);
for (let [name, c] of Object.entries(h)) {
  const hashToCurve = c.hashToCurve(msg).toHex();
  const hashToCurve_customDST = c.hashToCurve(msg, { DST: 'hello noble' }).toHex();
  const encodeToCurve = 'encodeToCurve' in c ? c.encodeToCurve(msg).toHex() : undefined;
  // ristretto255, decaf448 only
  const deriveToCurve = 'deriveToCurve' in c ?
    c.deriveToCurve!(new Uint8Array(c.Point.Fp.BYTES * 2)).toHex() : undefined;
  const hashToScalar = c.hashToScalar(msg);
  console.log({
    name, hashToCurve, hashToCurve_customDST, encodeToCurve, deriveToCurve, hashToScalar
  });
}

// abstract methods
import { expand_message_xmd, expand_message_xof, hash_to_field } from '@noble/curves/abstract/hash-to-curve.js';

The module allows to hash arbitrary strings to elliptic curve points. Implements RFC 9380.

[!NOTE] Why is p256_hasher separate from p256? The methods reside in separate _hasher namespace for tree-shaking: this way users who don't need hash-to-curve, won't have it in their builds.

OPRFs

import { p256_oprf, p384_oprf, p521_oprf } from '@noble/curves/nist.js';
import { ristretto255_oprf } from '@noble/curves/ed25519.js';
import { decaf448_oprf } from '@noble/curves/ed448.js';

We provide OPRFs (oblivious pseudorandom functions), conforming to RFC 9497.

OPRF allows to interactively create an Output = PRF(Input, serverSecretKey):

• Server cannot calculate Output by itself: it doesn't know Input
• Client cannot calculate Output by itself: it doesn't know server secretKey
• An attacker interception the communication can't restore Input/Output/serverSecretKey and can't link Input to some value.

FROST threshold signatures

FROST implements RFC 9591 threshold Schnorr signing. It is similar to multisig from the application point of view: any min of max participants can jointly produce one Schnorr signature under a shared public key. Supported ciphersuites are p256_FROST, ed25519_FROST, ed448_FROST, ristretto255_FROST, secp256k1_FROST, and schnorr_FROST (Taproot-compatible secp256k1). Signing has two rounds: selected signers commit first, then produce signature shares.

[!WARNING] The FROST code is new and has not been audited yet. It passes the imported frost-rs vectors/examples and local regression tests.

import { p256_FROST } from '@noble/curves/nist.js';

const signers = { min: 2, max: 3 };
const alice = p256_FROST.Identifier.derive('alice@example.com');
const bob = p256_FROST.Identifier.derive('bob@example.com');
const carol = p256_FROST.Identifier.derive('carol@example.com');
// trusted dealer
const deal = p256_FROST.trustedDealer(signers, [alice, bob, carol]);
for (const id of [alice, bob, carol]) p256_FROST.validateSecret(deal.secretShares[id], deal.public);

const msg = new TextEncoder().encode('hello threshold');
// round 1: selected signers commit
const aliceRound1 = p256_FROST.commit(deal.secretShares[alice]);
const bobRound1 = p256_FROST.commit(deal.secretShares[bob]);
const commitmentList = [aliceRound1.commitments, bobRound1.commitments];
// round 2: signers produce signature shares
const sigShares = {
  [alice]: p256_FROST.signShare(
    deal.secretShares[alice],
    deal.public,
    aliceRound1.nonces,
    commitmentList,
    msg
  ),
  [bob]: p256_FROST.signShare(
    deal.secretShares[bob],
    deal.public,
    bobRound1.nonces,
    commitmentList,
    msg
  ),
};
const sig = p256_FROST.aggregate(deal.public, commitmentList, msg, sigShares);
const isValid = p256_FROST.verify(sig, msg, deal.public.commitments[0]);

Key generation can be done with a trusted dealer or with DKG. DKG has three rounds: participants commit to key generation, exchange private shares, then derive final participant keys.

import { p256_FROST } from '@noble/curves/nist.js';

const signers = { min: 2, max: 3 };
const alice = p256_FROST.DKG.round1(p256_FROST.Identifier.fromNumber(1), signers);
const bob = p256_FROST.DKG.round1(p256_FROST.Identifier.fromNumber(2), signers);
const carol = p256_FROST.DKG.round1(p256_FROST.Identifier.fromNumber(3), signers);

// round 1: participants commit to key generation
const aliceRound2 = p256_FROST.DKG.round2(alice.secret, [bob.public, carol.public]);
const bobRound2 = p256_FROST.DKG.round2(bob.secret, [alice.public, carol.public]);
const carolRound2 = p256_FROST.DKG.round2(carol.secret, [alice.public, bob.public]);

// round 2: private shares for each recipient
const aliceKey = p256_FROST.DKG.round3(alice.secret, [bob.public, carol.public], [
  bobRound2[p256_FROST.Identifier.fromNumber(1)],
  carolRound2[p256_FROST.Identifier.fromNumber(1)],
]);
// round 3: final participant key package

DKG helpers are intended for interoperable testing and practical key generation. The library implements the cryptographic steps, not the surrounding application protocol: callers still need authenticated communication, coordination, retries, session handling, and policy.

poseidon: Poseidon hash

Implements Poseidon ZK-friendly hash: permutation and sponge.

There are many poseidon variants with different constants. We don't provide them: you should construct them manually. Check out scure-starknet package for a proper example.

import { bn254 } from '@noble/curves/bn254.js';
import { grainGenConstants, poseidon, poseidonSponge } from '@noble/curves/abstract/poseidon.js';

const rate = 2;
const capacity = 1;
const Fp = bn254.fields.Fr;
const { mds, roundConstants } = grainGenConstants({
  Fp,
  t: rate + capacity,
  roundsFull: 8,
  roundsPartial: 31,
});
const opts = {
  Fp,
  rate,
  capacity,
  sboxPower: 17,
  mds,
  roundConstants,
  roundsFull: 8,
  roundsPartial: 31,
};
const permutation = poseidon({ ...opts, t: rate + capacity });
const sponge = poseidonSponge(opts); // use carefully, not specced

fft: Fast Fourier Transform

import * as fft from '@noble/curves/abstract/fft.js';
import { bls12_381 } from '@noble/curves/bls12-381.js';
const Fr = bls12_381.fields.Fr;
const roots = fft.rootsOfUnity(Fr, 7n);
const fftFr = fft.FFT(roots, Fr);

Experimental implementation of NTT / FFT (Fast Fourier Transform) over finite fields. API may change at any time. The code has not been audited. Feature requests are welcome.

utils: byte shuffling, conversion

import { bytesToHex, concatBytes, equalBytes, hexToBytes } from '@noble/curves/utils.js';

bytesToHex(Uint8Array.from([0xca, 0xfe, 0x01, 0x23]));
hexToBytes('cafe0123');
concatBytes(Uint8Array.from([0xca, 0xfe]), Uint8Array.from([0x01, 0x23]));
equalBytes(Uint8Array.of(0xca), Uint8Array.of(0xca));

Internals

Elliptic curve Point math

import { pippenger } from '@noble/curves/abstract/curve.js';
import { secp256k1, schnorr } from '@noble/curves/secp256k1.js';
import { p256, p384, p521 } from '@noble/curves/nist.js';
import { ed25519, ristretto255 } from '@noble/curves/ed25519.js';
import { ed448, decaf448 } from '@noble/curves/ed448.js';
import { bls12_381 } from '@noble/curves/bls12-381.js';
import { bn254 } from '@noble/curves/bn254.js';
import { jubjub, babyjubjub } from '@noble/curves/misc.js';

const curves = [
  secp256k1, schnorr, p256, p384, p521, ed25519, ed448,
  ristretto255, decaf448,
  bls12_381.G1, bls12_381.G2, bn254.G1,
  jubjub, babyjubjub
];
for (const curve of curves) {
  const { Point } = curve;
  const { BASE, ZERO, Fp, Fn } = Point;
  const info = Point.CURVE?.();
  const p = BASE.multiply(2n);

  // Initialization
  if (info?.type === 'weierstrass') {
    // projective (homogeneous) coordinates: (X, Y, Z) ∋ (x=X/Z, y=Y/Z)
    const p_ = new Point(BASE.X, BASE.Y, BASE.Z);
  } else if (info?.type === 'edwards') {
    // extended coordinates: (X, Y, Z, T) ∋ (x=X/Z, y=Y/Z)
    const p_ = new Point(BASE.X, BASE.Y, BASE.Z, BASE.T);
  }

  // Math
  const p1 = p.add(p);
  const p2 = p.double();
  const p3 = p.subtract(p);
  const p4 = p.negate();
  const p5 = p.multiply(451n);

  // MSM (multi-scalar multiplication)
  const pa = [BASE, BASE.multiply(2n), BASE.multiply(4n), BASE.multiply(8n)];
  const p6 = pippenger(Point, pa, [3n, 5n, 7n, 11n]);
  const _true3 = p6.equals(BASE.multiply(129n)); // 129*G

  const pcl = p.clearCofactor();
  const isTorsionFree = p.isTorsionFree();

  const r1 = p.toBytes();
  const r1_ = Point.fromBytes(r1);
  const r2 = p.toAffine();
  const { x, y } = r2;
  const r2_ = Point.fromAffine(r2);
}

modular: Modular arithmetics & finite fields

import { mod, invert, Field } from '@noble/curves/abstract/modular.js';

// Finite Field utils
const fp = Field(2n ** 255n - 19n); // Finite field over 2^255-19
fp.mul(591n, 932n); // multiplication
fp.pow(481n, 11024858120n); // exponentiation
fp.div(5n, 17n); // division: 5/17 mod 2^255-19 == 5 * invert(17)
fp.inv(5n); // modular inverse
fp.sqrt(4n); // square root

// Non-Field generic utils are also available
mod(21n, 10n); // 21 mod 10 == 1n; fixed version of 21 % 10
invert(17n, 10n); // invert(17) mod 10; modular multiplicative inverse

All arithmetics is done with JS bigints over finite fields, which is defined from modular sub-module.

Field operations are not constant-time: see security. The fact is mostly irrelevant, but the important method to keep in mind is pow, which may leak exponent bits, when used naïvely.

weierstrass: Custom Weierstrass curve

import { weierstrass } from '@noble/curves/abstract/weierstrass.js';
// NIST secp192r1 aka p192. https://www.secg.org/sec2-v2.pdf
const p192_CURVE = {
  p: 0xfffffffffffffffffffffffffffffffeffffffffffffffffn,
  n: 0xffffffffffffffffffffffff99def836146bc9b1b4d22831n,
  h: 1n,
  a: 0xfffffffffffffffffffffffffffffffefffffffffffffffcn,
  b: 0x64210519e59c80e70fa7e9ab72243049feb8deecc146b9b1n,
  Gx: 0x188da80eb03090f67cbf20eb43a18800f4ff0afd82ff1012n,
  Gy: 0x07192b95ffc8da78631011ed6b24cdd573f977a11e794811n,
};
const p192_Point = weierstrass(p192_CURVE);

Short Weierstrass curve's formula is y² = x³ + ax + b. weierstrass expects arguments a, b, field characteristic p, curve order n, cofactor h and coordinates Gx, Gy of generator point.

edwards: Custom Edwards curve

import { edwards } from '@noble/curves/abstract/edwards.js';
const ed25519_CURVE = {
  p: 0x7fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffedn,
  n: 0x1000000000000000000000000000000014def9dea2f79cd65812631a5cf5d3edn,
  h: 8n,
  a: 0x7fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffecn,
  d: 0x52036cee2b6ffe738cc740797779e89800700a4d4141d8ab75eb4dca135978a3n,
  Gx: 0x216936d3cd6e53fec0a4e231fdd6dc5c692cc7609525a7b2c9562d608f25d51an,
  Gy: 0x6666666666666666666666666666666666666666666666666666666666666658n,
};
const ed25519_Point = edwards(ed25519_CURVE);

Twisted Edwards curve's formula is ax² + y² = 1 + dx²y². You must specify a, d, field characteristic p, curve order n (sometimes named as L), cofactor h and coordinates Gx, Gy of generator point.

Custom ECDSA instance

import { weierstrass, ecdsa } from '@noble/curves/abstract/weierstrass.js';
import { sha224, sha256 } from '@noble/hashes/sha2.js';
const p192_CURVE = {
  p: 0xfffffffffffffffffffffffffffffffeffffffffffffffffn,
  n: 0xffffffffffffffffffffffff99def836146bc9b1b4d22831n,
  h: 1n,
  a: 0xfffffffffffffffffffffffffffffffefffffffffffffffcn,
  b: 0x64210519e59c80e70fa7e9ab72243049feb8deecc146b9b1n,
  Gx: 0x188da80eb03090f67cbf20eb43a18800f4ff0afd82ff1012n,
  Gy: 0x07192b95ffc8da78631011ed6b24cdd573f977a11e794811n,
};
const p192_Point = weierstrass(p192_CURVE);
const p192_sha256 = ecdsa(p192_Point, sha256);
// or
const p192_sha224 = ecdsa(p192_Point, sha224);

const keys = p192_sha256.keygen();
const msg = new TextEncoder().encode('custom curve');
const sig = p192_sha256.sign(msg, keys.secretKey);
const isValid = p192_sha256.verify(sig, msg, keys.publicKey);

Security

The library has been audited:

• at version 2.2.0, in Apr 2026, by ourselves (self-audited)
  • Scope: everything
  • Changes since audit
• at version 1.6.0, in Sep 2024, independently, by Cure53
  • PDFs: website, in-repo
  • Changes since audit
  • Scope: ed25519, ed448, their add-ons, bls12-381, bn254, hash-to-curve, low-level primitives bls, tower, edwards, montgomery.
  • The audit has been funded by OpenSats
• at version 1.2.0, in Sep 2023, independently, by Kudelski Security
  • PDFs: in-repo
  • Changes since audit
  • Scope: scure-starknet and its related abstract modules of noble-curves: curve, modular, poseidon, weierstrass
  • The audit has been funded by Starkware
• at version 0.7.3, in Feb 2023, independently, by Trail of Bits
  • PDFs: website, in-repo
  • Changes since audit
  • Scope: abstract modules curve, hash-to-curve, modular, poseidon, utils, weierstrass and top-level modules _shortw_utils and secp256k1
  • The audit has been funded by Ryan Shea

It is tested against property-based, cross-library and Wycheproof vectors, and is being fuzzed in the separate repo.

If you see anything unusual: investigate and report.

Constant-timeness

We're targetting algorithmic constant time. JIT-compiler and Garbage Collector make "constant time" extremely hard to achieve timing attack resistance in a scripting language. Which means any other JS library can't have constant-timeness. Even statically typed Rust, a language without GC, makes it harder to achieve constant-time for some cases. If your goal is absolute security, don't use any JS lib — including bindings to native ones. Use low-level libraries & languages.

Memory dumping

Use low-level languages instead of JS / WASM if your goal is absolute security.

The library mostly uses Uint8Arrays and bigints.

• Uint8Arrays have .fill(0) which instructs to fill content with zeroes but there are no guarantees in JS
• bigints are immutable and don't have a method to zeroize their content: a user needs to wait until the next garbage collection cycle
• hex strings are also immutable: there is no way to zeroize them
• await fn() will write all internal variables to memory. With async functions there are no guarantees when the code chunk would be executed. Which means attacker can have plenty of time to read data from memory.

This means some secrets could stay in memory longer than anticipated. However, if an attacker can read application memory, it's doomed anyway: there is no way to guarantee anything about zeroizing sensitive data without complex tests-suite which will dump process memory and verify that there is no sensitive data left. For JS it means testing all browsers (including mobile). And, of course, it will be useless without using the same test-suite in the actual application that consumes the library.

Supply chain security

• Commits are signed with PGP keys to prevent forgery. Be sure to verify the commit signatures
• Releases are made transparently through token-less GitHub CI and Trusted Publishing. Be sure to verify the provenance logs for authenticity.
• Rare releasing is practiced to minimize the need for re-audits by end-users.
• Dependencies are minimized and strictly pinned to reduce supply-chain risk.
  • We use as few dependencies as possible.
  • Version ranges are locked, and changes are checked with npm-diff.
• Dev dependencies are excluded from end-user installs; they’re only used for development and build steps.

For this package, there is 1 dependency; and a few dev dependencies:

• noble-hashes provides cryptographic hashing functionality
• jsbt is used for benchmarking / testing / build tooling and developed by the same author
• prettier, fast-check and typescript are used for code quality / test generation / ts compilation

Randomness

We rely on the built-in crypto.getRandomValues, which is considered a cryptographically secure PRNG.

Browsers have had weaknesses in the past - and could again - but implementing a userspace CSPRNG is even worse, as there’s no reliable userspace source of high-quality entropy.

Quantum computers

Cryptographically relevant quantum computer, if built, will allow to break elliptic curve cryptography (both ECDSA / EdDSA & ECDH) using Shor's algorithm.

Consider switching to newer / hybrid algorithms, such as SPHINCS+. They are available in noble-post-quantum.

NIST prohibits classical cryptography (RSA, DSA, ECDSA, ECDH) after 2035. Australian ASD prohibits it after 2030.

Speed

npm run bench

noble-curves spends 10+ ms to generate 20MB+ of base point precomputes. This is done one-time per curve.

The generation is deferred until any method (pubkey, sign, verify) is called. User can force precompute generation by manually calling Point.BASE.precompute(windowSize, false). Check out the source code.

Benchmark results on Apple M4:

# secp256k1
init 10ms
getPublicKey x 9,099 ops/sec @ 109μs/op
sign x 7,182 ops/sec @ 139μs/op
verify x 1,188 ops/sec @ 841μs/op
recoverPublicKey x 1,265 ops/sec @ 790μs/op
getSharedSecret x 735 ops/sec @ 1ms/op
schnorr.sign x 957 ops/sec @ 1ms/op
schnorr.verify x 1,210 ops/sec @ 825μs/op

# ed25519
init 14ms
getPublicKey x 14,216 ops/sec @ 70μs/op
sign x 6,849 ops/sec @ 145μs/op
verify x 1,400 ops/sec @ 713μs/op

# ed448
init 37ms
getPublicKey x 5,273 ops/sec @ 189μs/op
sign x 2,494 ops/sec @ 400μs/op
verify x 476 ops/sec @ 2ms/op

# p256
init 17ms
getPublicKey x 8,977 ops/sec @ 111μs/op
sign x 7,236 ops/sec @ 138μs/op
verify x 877 ops/sec @ 1ms/op

# p384
init 42ms
getPublicKey x 4,084 ops/sec @ 244μs/op
sign x 3,247 ops/sec @ 307μs/op
verify x 331 ops/sec @ 3ms/op

# p521
init 83ms
getPublicKey x 2,049 ops/sec @ 487μs/op
sign x 1,748 ops/sec @ 571μs/op
verify x 170 ops/sec @ 5ms/op

# ristretto255
add x 931,966 ops/sec @ 1μs/op
multiply x 15,444 ops/sec @ 64μs/op
encode x 21,367 ops/sec @ 46μs/op
decode x 21,715 ops/sec @ 46μs/op

# decaf448
add x 478,011 ops/sec @ 2μs/op
multiply x 416 ops/sec @ 2ms/op
encode x 8,562 ops/sec @ 116μs/op
decode x 8,636 ops/sec @ 115μs/op

# ECDH
x25519 x 1,981 ops/sec @ 504μs/op
x448 x 743 ops/sec @ 1ms/op
secp256k1 x 728 ops/sec @ 1ms/op
p256 x 705 ops/sec @ 1ms/op
p384 x 268 ops/sec @ 3ms/op
p521 x 137 ops/sec @ 7ms/op

# hash-to-curve
hashToPrivateScalar x 1,754,385 ops/sec @ 570ns/op
hash_to_field x 135,703 ops/sec @ 7μs/op
hashToCurve secp256k1 x 3,194 ops/sec @ 313μs/op
hashToCurve p256 x 5,962 ops/sec @ 167μs/op
hashToCurve p384 x 2,230 ops/sec @ 448μs/op
hashToCurve p521 x 1,063 ops/sec @ 940μs/op
hashToCurve ed25519 x 4,047 ops/sec @ 247μs/op
hashToCurve ed448 x 1,691 ops/sec @ 591μs/op
hash_to_ristretto255 x 8,733 ops/sec @ 114μs/op
hash_to_decaf448 x 3,882 ops/sec @ 257μs/op

# modular over secp256k1 P field
invert a x 866,551 ops/sec @ 1μs/op
invert b x 693,962 ops/sec @ 1μs/op
sqrt p = 3 mod 4 x 25,738 ops/sec @ 38μs/op
sqrt tonneli-shanks x 847 ops/sec @ 1ms/op

# bls12-381
init 22ms
getPublicKey x 1,325 ops/sec @ 754μs/op
sign x 80 ops/sec @ 12ms/op
verify x 62 ops/sec @ 15ms/op
pairing x 166 ops/sec @ 6ms/op
pairing10 x 54 ops/sec @ 18ms/op ± 23.48% (15ms..36ms)
MSM 4096 scalars x points 3286ms
aggregatePublicKeys/8 x 173 ops/sec @ 5ms/op
aggregatePublicKeys/32 x 46 ops/sec @ 21ms/op
aggregatePublicKeys/128 x 11 ops/sec @ 84ms/op
aggregatePublicKeys/512 x 2 ops/sec @ 335ms/op
aggregatePublicKeys/2048 x 0 ops/sec @ 1346ms/op
aggregateSignatures/8 x 82 ops/sec @ 12ms/op
aggregateSignatures/32 x 21 ops/sec @ 45ms/op
aggregateSignatures/128 x 5 ops/sec @ 178ms/op
aggregateSignatures/512 x 1 ops/sec @ 705ms/op
aggregateSignatures/2048 x 0 ops/sec @ 2823ms/op

Upgrading

Supported node.js versions:

• v2 (2025-08): v20.19+ (ESM-only)
• v1 (2023-04): v14.21+ (ESM & CJS)

Changelog of curves v1 to curves v2

v2 massively simplifies internals, improves security, reduces bundle size and lays path for the future. We tried to keep v2 as much backwards-compatible as possible.

To simplify upgrading, upgrade first to curves 1.9.x. It would show deprecations in vscode-like text editor. Fix them first.

• The package is now ESM-only. ESM can finally be loaded from common.js on node v20.19+
• .js extension must be used for all modules
  • Old: @noble/curves/ed25519
  • New: @noble/curves/ed25519.js
  • This simplifies working in browsers natively without transpilers

New features:

• webcrypto: create friendly noble-like wrapper over built-in WebCrypto
• oprf: implement RFC 9497 OPRFs (oblivious pseudorandom functions)
  • We support p256, p384, p521, ristretto255 and decaf448
• weierstrass, edwards: add isValidSecretKey, isValidPublicKey
• misc: add Brainpool curves: brainpoolP256r1, brainpoolP384r1, brainpoolP512r1

Changes:

• Most methods now expect Uint8Array, string hex inputs are prohibited
  • The change simplifies reasoning, improves security and reduces malleability
  • Point.fromHex now expects string-only hex inputs, use Point.fromBytes for Uint8Array
• Many methods were renamed, upgrade to curves v1.9 first to highlight deprecated old names
• Breaking changes of ECDSA (secp256k1, p256, p384...):
  • To bring back old behavior, pass { prehash: false, lowS: false } to sign / verify
  • sign, verify: Switch to prehashed messages. Instead of messageHash, the methods now expect unhashed message. To bring back old behavior, use option {prehash: false}
  • sign, verify: Switch to lowS signatures by default. This change doesn't affect secp256k1, which has been using lowS since beginning. To bring back old behavior, use option {lowS: false}
  • sign, verify: Switch to Uint8Array signatures (format: 'compact') by default.
  • verify: der format must be explicitly specified in {format: 'der'}. This reduces malleability
  • verify: prohibit Signature-instance signature. User must now always do signature.toBytes()
• Breaking changes of BLS signatures (bls12-381, bn254):
  • Move getPublicKey, sign, verify, signShortSignature etc into two new namespaces: bls.longSignatures (G1 pubkeys, G2 sigs) and bls.shortSignatures (G1 sigs, G2 pubkeys).
  • verifyBatch now expects array of inputs {message: ..., publicKey: ...}[]
• Curve changes:
  • Massively simplify curve creation, split it into point creation & sig generator creation
  • New methods are weierstrass() + ecdsa() / edwards() + eddsa()
  • weierstrass / edwards expect simplified curve params (Fp became p)
  • ecdsa / eddsa expect Point class and hash
  • Remove unnecessary Fn argument in pippenger
• modular changes:
  • Field#fromBytes() now validates elements to be in 0..order-1 range
• Massively improve error messages, make them more descriptive

Renamings:

• Module changes
  • p256, p384, p521 modules have been moved into nist
  • jubjub module has been moved into misc
• Point changes
  • ExtendedPoint, ProjectivePoint => Point
  • Point coordinates (projective / extended) from px/ex, py/ey, pz/ez, et => X, Y, Z, T
  • Point.normalizeZ, Point.msm => separate methods in abstract/curve.js submodule
  • Point.fromPrivateKey() got removed, use Point.BASE.multiply() and Point.Fn.fromBytes(secretKey)
  • toRawBytes, fromRawBytes => toBytes, fromBytes
  • RistrettoPoint => ristretto255.Point, DecafPoiont => decaf448.Point
• Signature (ECDSA) changes
  • toCompactRawBytes, toDERRawBytes => toBytes('compact'), toBytes('der')
  • toCompactHex, toDERHex => toHex('compact'), toHex('der')
  • fromCompact, fromDER => fromBytes(format), fromHex(format)
• utils changes
  • randomPrivateKey => randomSecretKey
  • utils.precompute, Point#_setWindowSize => Point#precompute
  • edwardsToMontgomery => utils.toMontgomery
  • edwardsToMontgomeryPriv => utils.toMontgomerySecret
• Rename all curve-specific hash-to-curve methods to *curve*_hasher. Example: secp256k1.hashToCurve => secp256k1_hasher.hashToCurve()
• Massive type renamings and improvements

Removed features:

• Point#multiplyAndAddUnsafe, Point#hasEvenY
• CURVE property with all kinds of random stuff. Point.CURVE() now replaces it, but only provides curve parameters
• Remove pasta, bn254_weierstrass (NOT pairing-based bn254) curves
• utils.normPrivateKeyToScalar - use Point.Fn.fromBytes
• Field.MASK

secp256k1 v1, ed25519 v1, bls12-381 v1 to curves v1

Previously, the library was split into single-feature packages noble-secp256k1, noble-ed25519 and noble-bls12-381.

Curves continue their original work. The single-feature packages changed their direction towards providing minimal 5kb implementations of cryptography, which means they have less features. Separate bls package had been deprecated.

secp256k1:

• getPublicKey: defaults to compressed sigs (use second argument to adjust)
• sign: renamed options canonical => lowS; der => format: 'der'
• verify: renamed option strict => lowS
• getSharedSecret: defaults to compressed sigs (use third argument to adjust)
• recoverPublicKey(msg, sig, rec) was changed to sig.recoverPublicKey(msg)
• Point (2d xy) has been changed to ProjectivePoint (3d xyz)

ed25519:

• Signature was removed in favor of raw bytes
• getSharedSecret was moved to x25519 module

bls12-381:

• Renamed PointG1 -> G1.Point, PointG2 -> G2.Point
• Renamed PointG2.fromSignature -> Signature.decode, PointG2.toSignature -> Signature.encode

Contributing & testing

• npm install && npm run build && npm test will build the code and run tests.
• npm run lint / npm run format will run linter / fix linter issues.
• npm run bench will run benchmarks
• npm run build:release will build single file

See paulmillr.com/noble for useful resources, articles, documentation and demos related to the library.

MuSig2 signature scheme and BIP324 ElligatorSwift mapping for secp256k1 are available in a separate package.

License

The MIT License (MIT)

Copyright (c) 2022 Paul Miller (https://paulmillr.com)

See LICENSE file.