Build Workflow License

stochastic-rs

A high-performance Rust library for stochastic process simulation, quantitative finance, statistics, copulas, distributions, and neural-network volatility surrogates. Generic over f32 / f64, with SIMD acceleration on CPU and CUDA / Metal / Accelerate / cubecl backends where they pay off, and first-class Python bindings via PyO3.

Documentation

📖 stochastic.rust-dd.com — full docs site (Fumadocs + Next.js, deployed on Vercel).

Highlights:

120+ stochastic processes — diffusion, jump, fractional / rough, short-rate, HJM, LMM, fBM, Hawkes, Lévy. Generic-precision ProcessExt<T> impl, SIMD on CPU, optional CUDA / Metal for FGN / fBM.
Pricing & calibration — closed-form (BSM, Bachelier, Black76, Bjerksund-Stensland, …), Fourier (Heston / Bates / Merton-jump / Kou / VG / CGMY / HKDE / double-Heston), Monte Carlo (basket, rainbow, cliquet, autocallable, spread), finite difference, Bermudan LSM, Heston SLV. Heston / SABR / SVJ / Lévy / rough Bergomi / double-Heston / Hull-White swaption-grid calibrators.
Statistics & risk — Hurst (Fukasawa), MLE for 1-D diffusions with 6 transition densities, ADF / KPSS / Phillips-Perron, realised variance with BNHLS bandwidth, HMM, changepoint, particle filter, UKF. VaR / CVaR / drawdown, Sharpe / Sortino / IR / Calmar.
Fixed income & credit — yield-curve bootstrapping, Nelson-Siegel / Svensson, multi-curve, IRS / inflation swaps, Vasicek / CIR / Hull-White / G2++ short-rate engines, Merton structural model, reduced-form survival curves, CDS pricing, JLT migration matrices.
Microstructure — Almgren-Chriss, Kyle (1985), Bouchaud propagator, full price-time priority order book.
Distributions & copulas — 19 SIMD distributions with closed-form pdf / cdf / cf / moments. Clayton / Frank / Gumbel / Independence bivariate; Gaussian / vine multivariate.
Python bindings — 210 entries (198 PyO3 classes + 12 functions) spanning every sub-crate except AI surrogates. Numpy-in / numpy-out.

Installation

Rust

[dependencies]
stochastic-rs = "2.0.0"

use stochastic_rs::prelude::*;
use stochastic_rs::stochastic::diffusion::gbm::Gbm;
use stochastic_rs::quant::pricing::heston::HestonPricer;

For per-sub-crate (lean) builds, OpenBLAS / CUDA / Metal / cubecl / Accelerate feature flags, native CPU optimisation, and SIMD details, see the installation guide on the docs site.

Python

pip install stochastic-rs

Source build (requires the Rust toolchain):

pip install maturin
maturin develop --release --manifest-path stochastic-rs-py/Cargo.toml

Linux (x86_64 / aarch64) and macOS (arm64 / x86_64) wheels ship with the openblas feature on. The Windows wheel omits the 15 BLAS-backed classes; everything else (≈195 classes / 12 functions) works identically. See the Python bindings page for the parity table and the source-build path with vcpkg.

Quickstart

use stochastic_rs::prelude::*;
use stochastic_rs::simd_rng::Unseeded;
use stochastic_rs::stochastic::diffusion::ou::Ou;
use stochastic_rs::quant::pricing::heston::HestonPricer;

fn main() {
    // Mean-reverting Ornstein-Uhlenbeck path: Ou::new(theta, mu, sigma, n, x0, t, seed)
    let ou = Ou::<f64>::new(2.0, 0.0, 1.0, 1_000, Some(0.0), Some(1.0), Unseeded);
    let path = ou.sample();
    println!("OU path points: {}", path.len());

    // Heston (1993) European option, closed form. HestonPricer::new args:
    // s, v0, k, r, q, rho, kappa, theta, sigma, lambda, tau, eval, expiration
    let pricer = HestonPricer::new(
        100.0, 0.04, 100.0, 0.03, Some(0.0),
        -0.5, 2.0, 0.04, 0.3, Some(0.0),
        Some(1.0), None, None,
    );
    let (call, put) = pricer.calculate_call_put();
    println!("call={call:.4}, put={put:.4}");
}

import stochastic_rs as srs

# Mean-reverting OU path
p = srs.Ou(theta=2.0, mu=0.0, sigma=1.0, n=1000, x0=0.0, t=1.0)
path = p.sample()                       # numpy.ndarray, shape (1000,)

# Heston European call
pricer = srs.HestonPricer(
    s0=100, k=100, tau=1.0, r=0.03, q=0.0,
    v0=0.04, kappa=2.0, theta=0.04, sigma=0.3, rho=-0.5,
)
print("call =", pricer.price("call"))
g = pricer.greeks("call")
print(f"delta={g.delta:.4f}, vega={g.vega:.4f}")

More end-to-end recipes (Heston calibration, fBM Hurst estimation, vol-surface from quotes, Python interop) live in the tutorials section.

Benchmarks

FGN — CPU vs CUDA native (`f32`, H = 0.7)

cargo bench --features cuda-native --bench fgn_cuda_native

Single path:

n	CPU `sample`	CUDA `.on(Device::CudaNative).sample()`	Speedup
1,024	8.1 µs	46 µs	0.18×
4,096	35 µs	84 µs	0.42×
16,384	147 µs	110 µs	1.3×
65,536	850 µs	227 µs	3.7×

Batch:

n, m	CPU `sample_par`	CUDA `.on(Device::CudaNative).sample_par`	Speedup
4,096, 32	147 µs	117 µs	1.3×
4,096, 512	1.78 ms	2.37 ms	0.75×
65,536, 128	12.6 ms	10.5 ms	1.2×
65,536, 1 k	102 ms	93 ms	1.1×

CUDA wins for large n (≥ 16 k); CPU rayon dominates for medium n because of the GPU launch / transfer overhead.

Distribution sampling — `Normal` vs upstream `rand_distr`

Single-thread fill_slice, median of 7 runs (cargo bench --bench dist_multicore). Comparison column:

rand_distr + SimdRng — rand_distr::Normal consuming our SimdRng (same uniform stream, only the Normal algorithm differs).
rand_distr + rand::rng() — the out-of-box upstream pipeline.

n	`SimdNormal` (µs)	`rand_distr + SimdRng` (µs)	speedup	`rand_distr + rand::rng()` (µs)	speedup
4	0.008	0.013	1.73×	0.032	4.22×
8	0.014	0.026	1.78×	0.065	4.52×
16	0.029	0.051	1.79×	0.128	4.47×
64	0.109	0.208	1.90×	0.508	4.64×
256	0.432	0.840	1.94×	2.029	4.70×
4 096	6.975	13.176	1.89×	32.382	4.64×
65 536	113.458	212.406	1.87×	520.219	4.59×

Single-sample speedup vs prior release

Criterion dist.sample(rng) loop, vs the wide 1.3.0 baseline (cargo bench --bench distributions -- --baseline before):

distribution	f32 / large	f64 / large	f64 / small
`Uniform/simd`	−57% (≈ 2.3×)	−77% (≈ 4.4×)	−58% (≈ 2.4×)
`Normal/simd`	−51% (≈ 2.0×)	−75% (≈ 4.0×)	−63% (≈ 2.7×)
`Exp/simd N=64`	−3% (n.s.)	−73% (≈ 3.7×)	—
`LogNormal/simd`	−71% (≈ 3.4×)	−70% (≈ 3.4×)	−66% (≈ 2.9×)

Driven by SIMD u64→f64 / u32→f32 magic-number conversion in SimdRng (direct-write fill_uniform_f64 / fill_uniform_f32 APIs that skip the [f64; 8] return-by-value round-trip), fused Exp(λ) scaling inside fill_exp_scaled, and an 8-at-a-time main loop in fill_ziggurat so copy_from_slice inlines to stp stores instead of a memcpy call.

Opt-in: dual-stream RNG (`dual-stream-rng` feature)

[dependencies]
stochastic-rs = { version = "2.1", features = ["dual-stream-rng"] }

Unlocks SimdRngDual (two parallel xoshiro engines) and SimdNormalDual (Ziggurat unrolled 2× over the dual streams). Measured against the single-stream SimdNormal::fill_slice on Apple Silicon (cargo bench --bench dual_stream_compare --features dual-stream-rng):

n	single (`SimdNormal`)	dual (`SimdNormalDual`)	Δ
64	111.6 ns	105.5 ns	−5.5%
256	444.8 ns	418.3 ns	−6.0%
4 096	7.43 µs	6.60 µs	−11.2%
65 536	113.9 µs	106.6 µs	−6.4%
1 048 576	1.83 ms	1.70 ms	−6.8%

The win comes from hiding the 16 scalar kn / wn table-lookup latencies behind the second engine's xoshiro state update on a modern out-of-order core. Uniform fills are not bottlenecked on the engine so they see no speedup. Trade-off: SimdRngDual::from_seed does not reproduce SimdRng::from_seed's bit-exact sequence (statistical properties are identical and KS-validated).

Contributing

Contributions are welcome — bug reports, feature suggestions, or PRs. Open an issue or start a discussion on GitHub. Per-feature recipes (add-diffusion-process, adding-distribution, calibration-pattern, docs-writing, …) live under .claude/skills/.

License

MIT — see LICENSE.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

stochastic_rs-2.5.1.tar.gz (1.0 MB view details)

Uploaded Jul 19, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

stochastic_rs-2.5.1-pp311-pypy311_pp73-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 PyPymanylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 PyPymanylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp315-cp315t-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.15tmanylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp315-cp315t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.15tmanylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp315-cp315-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.15manylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp315-cp315-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.15manylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp314-cp314t-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.14tmanylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp314-cp314t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.14tmanylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp314-cp314-win_amd64.whl (3.4 MB view details)

Uploaded Jul 19, 2026 CPython 3.14Windows x86-64

stochastic_rs-2.5.1-cp314-cp314-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.14manylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp314-cp314-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.14manylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp314-cp314-macosx_14_0_arm64.whl (12.5 MB view details)

Uploaded Jul 19, 2026 CPython 3.14macOS 14.0+ ARM64

stochastic_rs-2.5.1-cp313-cp313-win_amd64.whl (3.4 MB view details)

Uploaded Jul 19, 2026 CPython 3.13Windows x86-64

stochastic_rs-2.5.1-cp313-cp313-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.13manylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.13manylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp313-cp313-macosx_14_0_arm64.whl (12.5 MB view details)

Uploaded Jul 19, 2026 CPython 3.13macOS 14.0+ ARM64

stochastic_rs-2.5.1-cp312-cp312-win_amd64.whl (3.4 MB view details)

Uploaded Jul 19, 2026 CPython 3.12Windows x86-64

stochastic_rs-2.5.1-cp312-cp312-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.12manylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.12manylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp312-cp312-macosx_14_0_arm64.whl (12.5 MB view details)

Uploaded Jul 19, 2026 CPython 3.12macOS 14.0+ ARM64

stochastic_rs-2.5.1-cp311-cp311-win_amd64.whl (3.4 MB view details)

Uploaded Jul 19, 2026 CPython 3.11Windows x86-64

stochastic_rs-2.5.1-cp311-cp311-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.11manylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.11manylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp311-cp311-macosx_14_0_arm64.whl (12.5 MB view details)

Uploaded Jul 19, 2026 CPython 3.11macOS 14.0+ ARM64

stochastic_rs-2.5.1-cp310-cp310-win_amd64.whl (3.4 MB view details)

Uploaded Jul 19, 2026 CPython 3.10Windows x86-64

stochastic_rs-2.5.1-cp310-cp310-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.10manylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.10manylinux: glibc 2.17+ x86-64

stochastic_rs-2.5.1-cp39-cp39-manylinux_2_28_aarch64.whl (8.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.9manylinux: glibc 2.28+ ARM64

stochastic_rs-2.5.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.2 MB view details)

Uploaded Jul 19, 2026 CPython 3.9manylinux: glibc 2.17+ x86-64

File details

Details for the file stochastic_rs-2.5.1.tar.gz.

File metadata

Download URL: stochastic_rs-2.5.1.tar.gz
Upload date: Jul 19, 2026
Size: 1.0 MB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: maturin/1.14.1

File hashes

Hashes for stochastic_rs-2.5.1.tar.gz
Algorithm	Hash digest
SHA256	`c6aadd25ccb1f9dccc631a8cdd380b2b87de02032fd8a1b5149d7e7c82e1d033`
MD5	`7ff392ad0fec61a154d1df6a5b0d4e84`
BLAKE2b-256	`ddda1dae86827ec8b27087f755faa9edf8f267978f380351ab51584e2a887b88`

Algorithm	Hash digest
SHA256	`ef04f5f97d1ff87eb54056bf9c536a0130c8af4f9fecc5dd57120ee7a0311ff2`
MD5	`3ad1a5f910bdaa8e957448936a766872`
BLAKE2b-256	`8dac5e7883564b29f616ad4c1bdaa0409760899779a2206bf395420c21d5dae3`

Algorithm	Hash digest
SHA256	`2a12929253658e7b39b78141c7ee4fc885fe8c25eb4bbf8b6d36c019b8623957`
MD5	`23f2c8f4806a56fe224f8b19be9c690d`
BLAKE2b-256	`5dedd062e62ed950c0178dc6b2dadd32b582bbcb0b9434bb97fc2115820bb7c8`

Algorithm	Hash digest
SHA256	`b7fe2a23c86bcf25e46d126506968ae3bd3e5c118e93a47867cb2eef7249513a`
MD5	`413ec0f8953fa70a537f67487aa2a03b`
BLAKE2b-256	`90cb122806463bf962c0537e3b8c5afb5595f26f9f1000e78a87aa6c7de8c502`

Algorithm	Hash digest
SHA256	`e8725c0c79bfdf01a9af0daed7bc6b366d10bba3e4fc703cf1d5e42cf87c85f3`
MD5	`079e8bcd473fbe84b6a43c342950a0b2`
BLAKE2b-256	`9ac367735ac73529363519d2a9c8ae8e8aff7d5f2be7f55992faab48b4a61bb1`

Algorithm	Hash digest
SHA256	`2b8d34a0d23d1df7e061c9c59b5ded80943c2b7c8ee78aa7b5166efedb90d40a`
MD5	`e1581817cf38ff7e2590ad6d4e602ecd`
BLAKE2b-256	`d8f04ead598ed5224cffb784f329a3f7ebdc692953f8f30e51a7e12e4566ad50`

Algorithm	Hash digest
SHA256	`38b8d6790f0f88bcc70f36b68e72ad5c9d7cef36bacd1eb8d1f28423a25d280c`
MD5	`c5bc0f23abaec0424aac96ebb0ca6c2d`
BLAKE2b-256	`9beb8fc25e1798cf0c5892654e1656d0539a43aa137eb67ad974c41d2dc7fdb5`

Algorithm	Hash digest
SHA256	`ad848d152154e2ae3ff0b93c00f2a8c3e639629023f602dd82b01c5f2beab909`
MD5	`09dbb77438c7149954e17403e3add598`
BLAKE2b-256	`e2964e476b5b96633b778a4e3e6e2fd165f32f388f8a305ddd310ba3962d8bcc`

Algorithm	Hash digest
SHA256	`b0057299341c70c2af90b6a2ebeb3180e1809c49482e842e845b626efa5ac847`
MD5	`e0ac83e8053a5793f288ea71a5f286ac`
BLAKE2b-256	`ce3d495ba1046900c564b09f78ad44a9268c0b9a0817ca0fcaaa2cf3c24eeddd`

Algorithm	Hash digest
SHA256	`0910eb7e877258cd6ff1e0ef205213044dbe4f8ba31cbcb927cb9bad5ce91ab3`
MD5	`8602fb06315b75c206c457d1784c10f5`
BLAKE2b-256	`8f968b1d4c83422772514682a1f3e98e6b2804f3ebd302a67d7ee6c2cb68c688`

Algorithm	Hash digest
SHA256	`2765929c44e270659cdca6923081e424d4672aad5d70ad78b6a22d4de59bb3ec`
MD5	`627ec911a519b6bd00fb51dcb7784ded`
BLAKE2b-256	`6acc7119c4c2cf96dfb3edeaf9bc00e50510964b7dc33315a3b729096f89755c`

Algorithm	Hash digest
SHA256	`203da0ecb91ecfd72aaad8949a2795b9314408c40c862bd8ee981c75961605a3`
MD5	`b7d3e40e33b6bc01b2fa3a3e6062d7ff`
BLAKE2b-256	`fc767e8a6d05f46d663dd9a4642b574ab49f1cf37f7845e997bd153a650afea0`

Algorithm	Hash digest
SHA256	`8bfac3681b350b1eb9c93d4f76f6eccf2bd8b3f2a16fbb5332de994ee4883fb1`
MD5	`7e9459cb3621d4a64b1eeb459527381f`
BLAKE2b-256	`c634adbec153347c9648cb75ad16c8b3d3c0d7175f3c89cd9d64c07588e6ad3a`

Algorithm	Hash digest
SHA256	`aa91cf63fa02ca1ab13af3046cdd41992354131dc10dacda6c1e25ab62166bc0`
MD5	`7ea8f61c64d1188c136e013f710bf666`
BLAKE2b-256	`a6c3bbd0b4e816290bbfc5afa6ed609c967b9c5305e755a075b8642648334e50`

Algorithm	Hash digest
SHA256	`1b8609eb77255de7d706108d3920c89da9f9921cd99797a6aa9bf1ce42d94c66`
MD5	`56a40f5f62f5251162b616b2997313ab`
BLAKE2b-256	`6a73d872cd0e65a912dafdf7e934b71e196bcec1b6534e9dc85de46dd0c9a58e`

Algorithm	Hash digest
SHA256	`c018674896d0901e36059eb7e3ad58d3e2549211817ed4fb1513fe14481662d7`
MD5	`e3e0f68119be211a430ce5b0e809aa9a`
BLAKE2b-256	`05ef88835948784947f7f6258154dfe85e8b7cc3bb6daac50400f9a074f21992`

Algorithm	Hash digest
SHA256	`498716716e8c22b1d6279b7114bb17e8efbdb0453ced85d10e234cae91613b4b`
MD5	`9da2e46e843105f37e7092348ecf1d87`
BLAKE2b-256	`74d8c99505a34c256646068a1dc2030bc0b5e752ffecb7da95258a88143a8fb3`

Algorithm	Hash digest
SHA256	`df9e40da454dbfbecdf1d2f1ad516f090a737d1f18584689bb5ff47e0a4946b0`
MD5	`837ff3f15c74770365755dcb4e0a6cf8`
BLAKE2b-256	`d782841c2466de9669dbc34fea9cb6cd1f437764c87e453a655eebdc77661e3e`

Algorithm	Hash digest
SHA256	`877e0d57ef305bd37e3cc26e8ac8325861077882e831d44709f61464d22e34c0`
MD5	`0b57c3a935cf89eca42ea950051b276d`
BLAKE2b-256	`88dae1858bb34b274977472fb7d30cb17cadededa1e2e715f2ba7fec27fd57ee`

Algorithm	Hash digest
SHA256	`4448a80403cfced95d33054bd2e89a87381ff30bbf1b9a84287a7a8cce3814e3`
MD5	`f09353eb70e047a312fb18c275e5196f`
BLAKE2b-256	`85a7db73b17d2657df36dc4a235b2320227bc4a5ef5cec4dfefb2859e83c1483`

Algorithm	Hash digest
SHA256	`b0e8f35dbf86a0ce3d3f4d50781ba36fc17b7b01ae366903f1b8f7a7a66bc660`
MD5	`b92844171f1ce5efc88003f33140c718`
BLAKE2b-256	`ed1586b34b51ecc1a0ff6dcb60896c1cc9227741c623940bcb5df73fa1625b31`

stochastic-rs 2.5.1

stochastic-rs

Documentation

Installation

Rust

Python

Quickstart

Benchmarks

FGN — CPU vs CUDA native (f32, H = 0.7)

Distribution sampling — Normal vs upstream rand_distr

Single-sample speedup vs prior release

Opt-in: dual-stream RNG (dual-stream-rng feature)

Contributing

License

Project links

Key dates

1 maintainer

License

Requires

Tags

Classifiers

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

FGN — CPU vs CUDA native (`f32`, H = 0.7)

Distribution sampling — `Normal` vs upstream `rand_distr`

Opt-in: dual-stream RNG (`dual-stream-rng` feature)

Algorithm	Hash digest
SHA256	`7eec1876e57fb4e7b81feaedadc3d213788798663120d0d4b0412f525c1d34b4`
MD5	`c7f2f7106434d6b62750de999b2cb3a7`
BLAKE2b-256	`bcb93868340bb604bdcc8623e097af7e60c71a53d69857af9a5d3becee9b2ff5`

Algorithm	Hash digest
SHA256	`a565684827ded325417c5a449554c68acda8a39b7422d91e6507ee2d3e0512a0`
MD5	`69949f1bacf4bb41971f362d2ec023f8`
BLAKE2b-256	`8cc79357613fef27c4c6769bcf69c7affd243eda7759d59ccda72c6365907e81`

Algorithm	Hash digest
SHA256	`7dd817ddb6b12288fbf246cbad2f175d228ab13d58031022c0d05281f864ef1c`
MD5	`68b6b29049de6bc939c7859b80a35473`
BLAKE2b-256	`bf4332372c8cea4d8acacce971c83a44b6734435a247ba19d96b80a6126734f1`

Algorithm	Hash digest
SHA256	`7a7bcc76015a5a72d72c4c5abd87becbcf0c6a08cc0dcb4b00107a71de044cb2`
MD5	`5b2cd4a3637db70b76963cc9c80f0719`
BLAKE2b-256	`d20c31a8f3fb91d56ebf997bed7bb5f71141ce22483536b14e499f60b2050ac1`

Algorithm	Hash digest
SHA256	`318d5a7bd410926967c5be49b52b96cbb59d3b9cfc914440422e13d51ea45fb0`
MD5	`ef8c38209c424ba5eafbb77655d72986`
BLAKE2b-256	`a35e4bcf2a5d97d44e0c67df6b43249b4b4cbfbf403740245edda0a68b703bc3`

Algorithm	Hash digest
SHA256	`425a4f6f0e86b1ad924d14da14dd02f0e9556b3aef8c8bac6df79021fce0c4c0`
MD5	`0b42c78050435829e6be6ec777b980ee`
BLAKE2b-256	`c1f01cf9811a0b5f30037f4cd25020c211f9765d7d000dc47788e4d5090d568c`

Algorithm	Hash digest
SHA256	`b90bce9bf05e9ab8a4be00538f8e27edb41dfdea7dc50eeedc0a124fd451d721`
MD5	`258400aa0c3c06a1f464a911e9472c2f`
BLAKE2b-256	`e2280ad3648bd2eea2b9117c9f73c982de0eb6752e3acedef151e2301c2133cd`

Algorithm	Hash digest
SHA256	`2085512df42a5b33f0fd18c7ede0568459ef1d8b40362455c08bd9739c1b4002`
MD5	`3e263808a78b6959a0fbafbe1f908f44`
BLAKE2b-256	`429695e937ff5df86c7ec83cb59908666c65ac51b595bdeeaf9b919f1dec1dd1`

Algorithm	Hash digest
SHA256	`79fe2a49da88a8d9c874ca88f870c5b7ebb9f072ad481b4a86c576f1f2348504`
MD5	`ca9e4b570f250eb1208c83ea3cfa1902`
BLAKE2b-256	`2785e72409e3b09e0296ed791ce7b9b25df51d722cba9bc24582fdceddf7db67`