Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/provablehq/snarkvm/llms.txt

Use this file to discover all available pages before exploring further.

​
About

The snarkvm-algorithms crate contains cryptographic algorithms and primitives for zero-knowledge proof construction in SnarkVM. This crate implements the mathematical foundations required for the Aleo blockchain’s proof system. Package: snarkvm-algorithms
Version: 4.4.0
License: Apache-2.0

​
Core Modules

Cryptographic Hashing

Poseidon and SHA-256 hash functions optimized for ZK circuits

SNARK Systems

Varuna zkSNARK implementation with AHP for R1CS

Polynomial Commitments

KZG10 and SonicKZG10 polynomial commitment schemes

FFT Operations

Fast Fourier Transform for polynomial arithmetic

​
Key Features

​
Zero-Knowledge Proofs

  • Varuna zkSNARK - Universal preprocessing zkSNARK with batch proving
  • AHP for R1CS - Algebraic Holographic Proof compilation
  • Fiat-Shamir transformation - Non-interactive proof generation

​
Polynomial Arithmetic

  • FFT/IFFT - O(n log n) polynomial operations
  • Evaluation domains - Powers-of-two roots of unity
  • Dense and sparse polynomials - Memory-efficient representations

​
Cryptographic Commitments

  • KZG polynomial commitments - Constant-size commitments with pairing-based verification
  • Batched opening proofs - Amortized verification cost
  • Degree bounds - Enforced polynomial degree constraints

​
Optimized Primitives

  • Multi-scalar multiplication - Pippenger’s algorithm with parallel execution
  • Poseidon hash - Algebraic hash function for efficient ZK circuits
  • Batch operations - Parallelized cryptographic operations

​
Module Organization

snarkvm-algorithms/
├── crypto_hash/        # Cryptographic hash functions
│   ├── poseidon.rs     # Poseidon hash and sponge
│   └── sha256.rs       # SHA-256 implementation
├── fft/                # Fast Fourier Transform
│   ├── domain.rs       # Evaluation domains
│   ├── evaluations.rs  # Lagrange evaluations
│   └── polynomial/     # Polynomial types
├── polycommit/         # Polynomial commitments
│   ├── kzg10/          # KZG10 scheme
│   └── sonic_pc/       # SonicKZG10 with batching
├── snark/              # zkSNARK implementations
│   └── varuna/         # Varuna proof system
├── msm/                # Multi-scalar multiplication
├── r1cs/               # R1CS constraint system
├── srs/                # Structured reference string
└── traits/             # Core trait definitions

​
Usage Example

use snarkvm_algorithms::{
    crypto_hash::Poseidon,
    fft::{DensePolynomial, EvaluationDomain},
    polycommit::kzg10::KZG10,
    snark::varuna::VarunaSNARK,
};

// Hash with Poseidon
let poseidon = Poseidon::<Fr, 4>::setup();
let hash = poseidon.evaluate(&[field1, field2, field3]);

// Polynomial operations
let domain = EvaluationDomain::new(256)?;
let poly = DensePolynomial::from_coefficients_vec(coeffs);
let evals = domain.fft(&poly);

// Polynomial commitments
let srs = KZG10::load_srs(max_degree)?;
let (commitment, randomness) = KZG10::commit(&powers, &poly, None, None)?;

// zkSNARK proving
type Varuna = VarunaSNARK<Bls12_377, PoseidonSponge<Fq, 2, 1>, DefaultMode>;
let proof = Varuna::prove(&prover, &fs_params, &pk, version, &circuit, rng)?;
let valid = Varuna::verify(&verifier, &fs_params, &vk, version, &input, &proof)?;

​
Performance

​
Parallelization

Most operations are parallelized using Rayon:
  • FFT operations split across available cores
  • MSM uses parallel bucket accumulation
  • Polynomial commitment batching
  • Parallel constraint synthesis

​
CUDA Acceleration

Enable the cuda feature for GPU acceleration of:
  • Multi-scalar multiplication (MSM)
  • FFT/IFFT operations
  • Polynomial evaluations
[dependencies]
snarkvm-algorithms = { version = "4.4.0", features = ["cuda"] }
  • snarkvm-fields - Finite field arithmetic implementations
  • snarkvm-curves - Elliptic curve group operations
  • snarkvm-console - High-level VM types
  • snarkvm-circuit - Circuit-level constraint synthesis

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us