Classical ML & Statistics — Are we learning yet?

ML toolkits & linear models partial

The scikit-learn-shaped shelf: the general-purpose toolkits you reach for first. What is maintained here is real, but the generation of toolkits a 2016 tutorial would have recommended is archived.

Featured: smartcore [crate · repo · docs]

The single-crate classical toolkit — linear and logistic regression, decision trees, ensembles, SVM, naive Bayes, k-nearest neighbors, clustering, decomposition, model selection — and, as of 2026, the faster-moving of Rust's two. After eighteen dead months it went 0.4.0 to 0.6.14 in 2026 with contributors beyond its maintainers, zero open pull requests, and a user guide that is actually kept current, which is the biggest status correction on this site. Pick it over linfa if you want one dependency and a working tutorial; pick linfa if you want to depend on one algorithm crate rather than all of them.

Machine Learning in Rust.

docsThe smartcore User Guide — the best classical-ML tutorial in Rust papersmartcore archived release with a citable DOI

version: 0.6.14
downloads: 150k / 90d
stars: 954
license: Apache-2.0
last commit: 2026-08-26
published: 2026-08-26
dependents: 24 owners · 7 repos

Featured: linfa [crate · repo · docs]

The modular scikit-learn-shaped toolkit: twenty linfa-* crates covering linear and logistic regression, elastic net, SVMs, decision trees, the new bagging/random-forest/AdaBoost ensembles, naive Bayes, k-means, DBSCAN, OPTICS, GMM, PCA, t-SNE, ICA and PLS. It was rescued from dormancy in January 2025 and has shipped 0.7 through 0.8.1 with outside contributors since. Be realistic about the pace: dozens of open issues and pull requests, and three offers of maintainership still unanswered on the tracker. Neither it nor smartcore replaces scikit-learn — there is no pipeline abstraction and no grid search anywhere in Rust.

A Machine Learning framework for Rust

docslinfa documentation postlinfa #367, Maintainership Discussion — open since 2025-01, and the honest status

version: 0.8.1
downloads: 468k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 13 owners · 22 repos

paiml/aprender [repo]

Next Generation Machine Learning, Statistics and Deep Learning in PURE Rust

stars: 120
last commit: 2026-09-14

New & experimental (5)

Collected, but not clearing the published listing bar — too new, too quiet, or with no use by anyone but its author that we can see. Plenty of good software is here, some of it years old: this says what we can measure, not what it is worth.

linregress [crate · repo · docs]

ordinary least squared linear regression with some basic statistics

version: 0.5.4
downloads: 885k / 90d
stars: 24
license: MIT
last commit: 2025-02-21
published: 2024-10-11
dependents: 13 owners · 1338 repos

linfa-linalg [crate · repo · docs]

Pure-Rust implementation of linear algebra routines for ndarray

version: 0.2.1
downloads: 534k / 90d
stars: 39
license: MIT/Apache-2.0
last commit: 2025-05-15
published: 2025-05-15
dependents: 11 owners · 13 repos

automl [crate · repo · docs]

Automated machine learning for classification and regression

version: 0.3.0
downloads: 197 / 90d
stars: 38
license: MIT OR Apache-2.0
last commit: 2025-11-16
published: 2025-11-15
dependents: 2 owners · 1 repo

ndarray-glm [crate · repo · docs]

Performs regression for generalized linear models using IRLS on data stored in arrays.

version: 0.1.0
downloads: 189 / 90d
stars: 35
license: MIT
last commit: 2026-03-02
published: 2026-03-02
dependents: 1 repo

rustyml [crate · repo · docs]

A high-performance machine learning & deep learning library in pure Rust, offering ML algorithms and neural network support

version: 0.15.0
downloads: 197 / 90d
stars: 342
license: MIT
last commit: 2026-09-10
published: 2026-08-21

Archive (5)

Kept so you stop looking for them. Archives are computed on every build and reverse themselves if the reason clears. Where a human has named the reason or a replacement, it cites the public issue behind that.

outbrain-inc/fwumious_wabbit [repo]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Fwumious Wabbit, fast on-line machine learning toolkit written in Rust

stars: 139
license: BSD-3-Clause
last commit: 2024-06-04

rusty-machine [crate · repo · docs]

Archived: repo archived. Detected automatically, and it un-archives itself if that changes.

A machine learning library.

version: 0.5.4
downloads: 5k / 90d
stars: 1.3k
license: MIT
last commit: 2020-02-14
published: 2017-01-06
dependents: 4 owners · 18 repos

rustlearn [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

A machine learning package for Rust.

version: 0.5.0
downloads: 1k / 90d
stars: 646
license: Apache-2.0
last commit: 2020-06-21
published: 2018-07-29
dependents: 3 owners · 6 repos

rustml [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

A library for doing maching learning in Rust.

version: 0.0.7
downloads: 141 / 90d
stars: 69
license: MIT/Apache-2.0
last commit: 2022-06-15
published: 2018-01-31
dependents: 1 owner

classifier [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Naive Bayes classifier

version: 0.0.3
downloads: 138 / 90d
stars: 31
license: Apache-2.0
last commit: 2015-04-11
published: 2015-12-11
dependents: 1 repo

Gradient boosting & trees partial

Decision trees, random forests and gradient-boosted ensembles. The capability is present in both pure-Rust and binding form; the XGBoost-bindings generation before it is archived.

Featured: smartcore [crate · repo · docs]

The single-crate classical toolkit — linear and logistic regression, decision trees, ensembles, SVM, naive Bayes, k-nearest neighbors, clustering, decomposition, model selection — and, as of 2026, the faster-moving of Rust's two. After eighteen dead months it went 0.4.0 to 0.6.14 in 2026 with contributors beyond its maintainers, zero open pull requests, and a user guide that is actually kept current, which is the biggest status correction on this site. Pick it over linfa if you want one dependency and a working tutorial; pick linfa if you want to depend on one algorithm crate rather than all of them.

Machine Learning in Rust.

docsThe smartcore User Guide — the best classical-ML tutorial in Rust papersmartcore archived release with a citable DOI

version: 0.6.14
downloads: 150k / 90d
stars: 954
license: Apache-2.0
last commit: 2026-08-26
published: 2026-08-26
dependents: 24 owners · 7 repos

Featured: linfa [crate · repo · docs]

The modular scikit-learn-shaped toolkit: twenty linfa-* crates covering linear and logistic regression, elastic net, SVMs, decision trees, the new bagging/random-forest/AdaBoost ensembles, naive Bayes, k-means, DBSCAN, OPTICS, GMM, PCA, t-SNE, ICA and PLS. It was rescued from dormancy in January 2025 and has shipped 0.7 through 0.8.1 with outside contributors since. Be realistic about the pace: dozens of open issues and pull requests, and three offers of maintainership still unanswered on the tracker. Neither it nor smartcore replaces scikit-learn — there is no pipeline abstraction and no grid search anywhere in Rust.

A Machine Learning framework for Rust

docslinfa documentation postlinfa #367, Maintainership Discussion — open since 2025-01, and the honest status

version: 0.8.1
downloads: 468k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 13 owners · 22 repos

Featured: perpetual [crate · repo · docs]

Hyperparameter-free gradient boosting: one budget knob instead of a search over learning rate, depth and estimators. It has a Rust core with Python and R bindings, and it ships the things that usually send people back to Python — SHAP values, ONNX and XGBoost export, monotonic constraints and drift monitoring. It is also the only working SHAP implementation in the Rust ecosystem, which is a comment on the interpretability shelf rather than on this crate.

A self-generalizing gradient boosting machine that doesn't need hyperparameter optimization

talkPerpetual: a hyperparameter-free gradient boosting machine (Scientific Computing in Rust 2024) postperpetual, This Week in Rust Crate of the Week (2024-09-25)

version: 3.0.0-rc.2
downloads: 2k / 90d
stars: 706
license: Apache-2.0
last commit: 2026-04-02
published: 2026-04-02

linfa-trees [crate · repo · docs]

A collection of tree-based algorithms

version: 0.8.1
downloads: 95k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 14 owners · 5 repos

Pushp-Kharat1/PkBoost [repo]

PKBoost: Adaptive GBDT for Concept Drift, Built from scratch in Rust, PKBoost manages changing data distributions in fraud detection with a fraud rate of 0.2%. It shows less than 2% degradation under drift. In comparison, XGBoost experiences a 31.8% drop and LightGBM a 42.5% drop

stars: 74
last commit: 2026-07-16

lightgbm3 [crate · repo · docs]

Rust bindings for LightGBM library

version: 1.0.11
downloads: 69k / 90d
stars: 59
license: MIT
last commit: 2026-09-08
published: 2026-09-08
dependents: 1 owner

linfa-ensemble [crate · repo · docs]

A general method for creating ensemble classifiers

version: 0.8.1
downloads: 7k / 90d
stars: 4.7k
license: MIT/Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 2 owners

catboost-rust [crate · repo · docs]

Rust bindings for CatBoost, a gradient boosting library for machine learning. Downloads CatBoost binaries at runtime for cross-platform compatibility.

version: 0.4.0
downloads: 6k / 90d
stars: 6
license: Apache-2.0
last commit: 2026-04-19
published: 2026-04-19
dependents: 1 owner

New & experimental (3)

Collected, but not clearing the published listing bar — too new, too quiet, or with no use by anyone but its author that we can see. Plenty of good software is here, some of it years old: this says what we can measure, not what it is worth.

gbdt [crate · repo · docs]

A implementation of Gradient Boosting Regression Tree in Rust programming language

version: 0.1.3
downloads: 13k / 90d
stars: 225
license: Apache-2.0
last commit: 2025-01-29
published: 2024-01-24
dependents: 5 owners · 1 repo

randomforest [crate · repo · docs]

Random forest regressor and classifier

version: 0.2.0
downloads: 333 / 90d
stars: 12
license: MIT
last commit: 2026-05-21
published: 2026-04-22
dependents: 1 owner · 1 repo

bosk [crate · repo · docs]

Pure-Rust LightGBM inference: parses the text model format directly — no FFI, zero deps. Optional ONNX Runtime and CatBoost backends behind the same small Model trait.

version: 0.1.0
downloads: 2k / 90d
stars: 2
license: MIT OR Apache-2.0
last commit: 2026-07-27
published: 2026-07-15

Archive (3)

Kept so you stop looking for them. Archives are computed on every build and reverse themselves if the reason clears. Where a human has named the reason or a replacement, it cites the public issue behind that.

xgboost [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Machine learning using XGBoost

version: 0.1.4
downloads: 514 / 90d
stars: 121
license: MIT
last commit: 2024-02-02
published: 2019-03-05
dependents: 1 owner · 1 repo

catboost-rs [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Unofficial Rust bindings for Catboost (Machine Learning Gradient Boosting Library from Yandex)

version: 0.1.8
downloads: 5k / 90d
stars: 11
license: MIT/Apache-2.0
last commit: 2022-10-21
published: 2022-10-21
dependents: 1 owner

forester [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

A crate for implementing various flavors of random forests and decision trees.

version: 0.0.3
downloads: 62 / 90d
stars: 12
license: MIT
last commit: 2021-08-04
published: 2018-06-30

Clustering, dim-reduction & nearest neighbors comprehensive

k-means, DBSCAN and HDBSCAN, dimensionality reduction, and the k-d and ball trees underneath them. Long-settled capability, and one of the few classical-ML shelves where you are spoiled for choice.

Featured: kiddo [crate · repo · docs]

A k-d tree that is fast enough and general enough to be infrastructure: float and fixed-point coordinates, arbitrary dimensionality, and immutable memory-mapped trees for indexes larger than RAM. It moves more downloads a quarter than any dedicated machine learning crate on this site. Use it for exact nearest neighbors in low dimensions — geometry, point clouds, spatial joins; for high-dimensional embedding search you want the approximate indexes on the retrieval hub instead.

A high-performance, flexible, ergonomic k-d tree library. Ideal for geo- and astro- nearest-neighbour and k-nearest-neighbor queries

benchmarkBosque, a rival k-d tree, drops its kiddo comparison and recommends kiddo v2 instead productionThe LSST/Rubin telescope-site M2 mirror controller depends on kiddo

version: 6.3.0
downloads: 945k / 90d
stars: 175
license: MIT OR Apache-2.0
last commit: 2026-09-06
published: 2026-09-06
dependents: 28 owners · 15 repos

Featured: smartcore [crate · repo · docs]

The single-crate classical toolkit — linear and logistic regression, decision trees, ensembles, SVM, naive Bayes, k-nearest neighbors, clustering, decomposition, model selection — and, as of 2026, the faster-moving of Rust's two. After eighteen dead months it went 0.4.0 to 0.6.14 in 2026 with contributors beyond its maintainers, zero open pull requests, and a user guide that is actually kept current, which is the biggest status correction on this site. Pick it over linfa if you want one dependency and a working tutorial; pick linfa if you want to depend on one algorithm crate rather than all of them.

Machine Learning in Rust.

docsThe smartcore User Guide — the best classical-ML tutorial in Rust papersmartcore archived release with a citable DOI

version: 0.6.14
downloads: 150k / 90d
stars: 954
license: Apache-2.0
last commit: 2026-08-26
published: 2026-08-26
dependents: 24 owners · 7 repos

Featured: linfa [crate · repo · docs]

The modular scikit-learn-shaped toolkit: twenty linfa-* crates covering linear and logistic regression, elastic net, SVMs, decision trees, the new bagging/random-forest/AdaBoost ensembles, naive Bayes, k-means, DBSCAN, OPTICS, GMM, PCA, t-SNE, ICA and PLS. It was rescued from dormancy in January 2025 and has shipped 0.7 through 0.8.1 with outside contributors since. Be realistic about the pace: dozens of open issues and pull requests, and three offers of maintainership still unanswered on the tracker. Neither it nor smartcore replaces scikit-learn — there is no pipeline abstraction and no grid search anywhere in Rust.

A Machine Learning framework for Rust

docslinfa documentation postlinfa #367, Maintainership Discussion — open since 2025-01, and the honest status

version: 0.8.1
downloads: 468k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 13 owners · 22 repos

kdtree [crate · repo · docs]

K-dimensional tree in Rust for fast geospatial indexing and nearest neighbors lookup

version: 0.8.1
downloads: 564k / 90d
stars: 284
license: MIT OR Apache-2.0
last commit: 2026-07-23
published: 2026-07-23
dependents: 34 owners · 49 repos

rstar [crate · repo · docs]

An R*-tree spatial index

version: 0.13.0
downloads: 13.4M / 90d
stars: 557
license: MIT OR Apache-2.0
last commit: 2026-09-11
published: 2026-05-24
dependents: 30 owners · 706 repos

linfa-clustering [crate · repo · docs]

A collection of clustering algorithms

version: 0.8.1
downloads: 203k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 25 owners · 6 repos

linfa-nn [crate · repo · docs]

A collection of nearest neighbour algorithms

version: 0.8.1
downloads: 310k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 19 owners · 12 repos

linfa-reduction [crate · repo · docs]

A collection of dimensionality reduction techniques

version: 0.8.1
downloads: 78k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 10 owners · 7 repos

kd-tree [crate · repo · docs]

k-dimensional tree

version: 0.6.2
downloads: 125k / 90d
stars: 86
license: MIT
last commit: 2025-11-16
published: 2025-11-16
dependents: 15 owners · 17 repos

bhtsne [crate · repo · docs]

Exact and Barnes-Hut implementations of t-SNE.

version: 0.7.13
downloads: 20k / 90d
stars: 95
license: MIT
last commit: 2026-08-31
published: 2026-08-31
dependents: 4 owners · 8 repos

petal-clustering [crate · repo · docs]

A collection of clustering algorithms.

version: 0.13.0
downloads: 26k / 90d
stars: 45
license: Apache-2.0
last commit: 2026-06-16
published: 2025-11-20
dependents: 4 owners

petal-neighbors [crate · repo · docs]

Nearest neighbor search algorithms including a ball tree and a vantage point tree.

version: 0.18.0
downloads: 53k / 90d
stars: 12
license: Apache-2.0
last commit: 2026-06-16
published: 2025-11-19
dependents: 4 owners · 1 repo

linfa-pls [crate · repo · docs]

Partial Least Squares family methods

version: 0.8.1
downloads: 14k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 2 owners · 1 repo

linfa-tsne [crate · repo · docs]

Barnes-Hut t-distributed stochastic neighbor embedding

version: 0.8.1
downloads: 7k / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23
dependents: 2 owners

annembed [crate · repo · docs]

a Rust implementation of a dimension reduction à la Umap

version: 0.1.6
downloads: 1k / 90d
stars: 45
license: MIT/Apache-2.0
last commit: 2026-02-28
published: 2025-11-19
dependents: 4 owners

hdbscan [crate · repo · docs]

HDBSCAN clustering in pure Rust. A huge improvement on DBSCAN, capable of identifying clusters of varying densities.

version: 0.12.0
downloads: 26k / 90d
stars: 40
license: MIT OR Apache-2.0
last commit: 2025-11-24
published: 2025-11-24
dependents: 6 owners

petal-decomposition [crate · repo · docs]

Matrix decomposition algorithms including PCA (principal component analysis) and ICA (independent component analysis)

version: 0.9.0
downloads: 1k / 90d
stars: 20
license: Apache-2.0
last commit: 2026-06-16
published: 2025-12-02
dependents: 1 owner · 1 repo

kmeans [crate · repo · docs]

Small and fast library for k-means clustering calculations.

version: 2.0.2
downloads: 1k / 90d
stars: 55
license: Apache-2.0
last commit: 2026-03-03
published: 2026-03-03
dependents: 2 owners · 2 repos

linfa-ica [crate · repo · docs]

A collection of Independent Component Analysis (ICA) algorithms

version: 0.8.1
downloads: 262 / 90d
stars: 4.7k
license: MIT OR Apache-2.0
last commit: 2026-08-22
published: 2025-12-23

New & experimental (3)

Collected, but not clearing the published listing bar — too new, too quiet, or with no use by anyone but its author that we can see. Plenty of good software is here, some of it years old: this says what we can measure, not what it is worth.

kmeans_colors [crate · repo · docs]

Simple k-means clustering to find dominant colors in images. Backed by a generic k-means implementation offered as a standalone library.

version: 0.7.1
downloads: 63k / 90d
stars: 173
license: MIT OR Apache-2.0
last commit: 2025-09-08
published: 2025-09-08
dependents: 11 owners · 4 repos

clustering [crate · repo · docs]

easy way to perform kmeans clustering on arbitrary data

version: 0.2.2
downloads: 1k / 90d
stars: 6
license: MIT
last commit: 2025-11-21
published: 2025-11-21
dependents: 2 owners · 1 repo

kmedoids [crate · repo · docs]

k-Medoids clustering with the FasterPAM algorithm

version: 0.5.5
downloads: 513 / 90d
stars: 30
license: GPL-3.0-or-later
last commit: 2026-05-22
published: 2026-05-22

Archive (3)

Kept so you stop looking for them. Archives are computed on every build and reverse themselves if the reason clears. Where a human has named the reason or a replacement, it cites the public issue behind that.

rusticsom [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Rust library for Self Organising Maps (SOM).

version: 1.1.1
downloads: 57 / 90d
stars: 36
license: MIT
last commit: 2020-10-07
published: 2020-04-01

cogset [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Generic implementations of clustering algorithms. Includes k-means, DBSCAN and OPTICS.

version: 0.2.0
downloads: 5k / 90d
stars: 21
license: MIT/Apache-2.0
last commit: 2015-10-13
published: 2015-12-15
dependents: 4 repos

kddbscan [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

A k -Deviation Density Based Clustering Algorithm (kDDBSCAN)

version: 0.1.0
downloads: 49 / 90d
stars: 5
license: MIT
last commit: 2020-06-13
published: 2020-06-13

Statistics partial

Distributions, summary statistics, hypothesis tests and correlations. The first two are well served in Rust; inferential statistics and regression diagnostics are not.

Featured: statrs [crate · repo · docs]

The foundation crate of statistics in Rust: distributions with pdf, cdf, inverse cdf and sampling, plus the statistical functions around them. It is the reason "do statistics in Rust" is not a caveat. For descriptive statistics over arrays reach for ndarray-stats, for streaming summaries average, and for Bayesian updating rv — statrs is the layer all three assume.

Statistical computing library for Rust

productionHugging Face's xet-core, the Xet transfer client, depends on statrs productionZama's tfhe-rs homomorphic encryption library depends on statrs

version: 0.19.1
downloads: 9.5M / 90d
stars: 827
license: MIT
last commit: 2026-09-07
published: 2026-08-11
dependents: 38 owners · 2285 repos

Featured: rv [crate · repo · docs]

Random variables done properly: distributions with conjugate priors and Bayesian updating, which is what separates it from statrs and makes it the base of most Bayesian work in Rust. It is the anchor of a shelf you assemble rather than adopt — pair it with changepoint for online changepoints, ganesh for MCMC, friedrich or egobox-gp for Gaussian processes, adskalman for filtering. What does not exist anywhere in Rust is a probabilistic programming language you would build a project on: every attempt is a solo project under fifty stars.

Random variables

productionGrafana's augurs builds its time-series outlier detection on rv

version: 0.20.1
downloads: 257k / 90d
stars: 37
license: MIT
last commit: 2026-05-21
published: 2026-05-21
dependents: 6 owners · 7 repos

average [crate · repo · docs]

Calculate statistics iteratively

version: 0.17.0
downloads: 1.2M / 90d
stars: 43
license: MIT OR Apache-2.0
last commit: 2026-04-15
published: 2026-04-15
dependents: 25 owners · 149 repos

special [crate · repo · docs]

The package provides special functions.

version: 0.14.1
downloads: 444k / 90d
stars: 20
license: Apache-2.0 OR MIT
last commit: 2026-08-08
published: 2026-08-08
dependents: 22 owners · 43 repos

ndarray-stats [crate · repo · docs]

Statistical routines for the n-dimensional array data structures provided by ndarray.

version: 0.7.0
downloads: 1.3M / 90d
stars: 237
license: MIT/Apache-2.0
last commit: 2025-12-28
published: 2025-12-28
dependents: 19 owners · 116 repos

peroxide [crate · repo · docs]

Rust comprehensive scientific computation library contains linear algebra, numerical analysis, statistics and machine learning tools with familiar syntax

version: 0.43.1
downloads: 162k / 90d
stars: 728
license: MIT OR Apache-2.0
last commit: 2026-08-10
published: 2026-08-02
dependents: 7 owners · 11 repos

rstats [crate · repo · docs]

Statistics, Information Measures, Data Analysis, Linear Algebra, Clifford Algebra, Machine Learning, Geometric Median, Matrix Decompositions, Mahalanobis Distance, Hulls, Multithreading..

version: 2.1.15
downloads: 6k / 90d
stars: 55
license: Apache-2.0
last commit: 2026-08-22
published: 2026-08-22
dependents: 6 owners · 5 repos

puruspe [crate · repo · docs]

Pure Rust Special function library

version: 0.4.4
downloads: 238k / 90d
stars: 29
license: MIT OR Apache-2.0
last commit: 2026-03-17
published: 2026-03-17
dependents: 9 owners · 42 repos

distrs [crate · repo · docs]

PDF, CDF, and percent-point/quantile functions for the normal and Student’s t distributions

version: 0.2.3
downloads: 176k / 90d
stars: 14
license: Unlicense OR MIT
last commit: 2026-04-07
published: 2026-04-07
dependents: 5 owners · 1 repo

hypors [crate · repo · docs]

Hypothesis Testing with Rust

version: 0.4.1
downloads: 2k / 90d
stars: 21
license: non-standard
last commit: 2026-09-13
published: 2026-09-13

New & experimental (4)

Collected, but not clearing the published listing bar — too new, too quiet, or with no use by anyone but its author that we can see. Plenty of good software is here, some of it years old: this says what we can measure, not what it is worth.

linregress [crate · repo · docs]

ordinary least squared linear regression with some basic statistics

version: 0.5.4
downloads: 885k / 90d
stars: 24
license: MIT
last commit: 2025-02-21
published: 2024-10-11
dependents: 13 owners · 1338 repos

watermill [crate · repo · docs]

Blazingly fast, generic, and serializable online statistics.

version: 0.1.2
downloads: 53k / 90d
stars: 76
license: MIT
last commit: 2025-07-31
published: 2025-07-12
dependents: 2 owners · 1 repo

incr_stats [crate · repo · docs]

Fast, scalable, incremental descriptive statistics in Rust

version: 1.0.2
downloads: 3k / 90d
stars: 8
license: MIT OR Apache-2.0
last commit: 2024-12-10
published: 2024-02-21
dependents: 2 owners

ndarray-glm [crate · repo · docs]

Performs regression for generalized linear models using IRLS on data stored in arrays.

version: 0.1.0
downloads: 189 / 90d
stars: 35
license: MIT
last commit: 2026-03-02
published: 2026-03-02
dependents: 1 repo

Archive (2)

Kept so you stop looking for them. Archives are computed on every build and reverse themselves if the reason clears. Where a human has named the reason or a replacement, it cites the public issue behind that.

medians [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Median, Statistical Measures, Mathematics, Statistics

version: 3.0.12
downloads: 27k / 90d
stars: 15
license: BSD-3-Clause
last commit: 2024-06-14
published: 2024-06-14
dependents: 3 owners · 2 repos

statest [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Rust crate for statistical test.

version: 0.2.2
downloads: 301k / 90d
stars: 7
license: MIT
last commit: 2021-02-24
published: 2021-02-24
dependents: 3 owners · 1 repo

Bayesian & probabilistic modeling partial

Bayesian and probabilistic modeling. This is a shelf of pieces — priors, samplers, filters, processes, changepoints — that you assemble yourself, rather than a framework you adopt.

Featured: rv [crate · repo · docs]

Random variables done properly: distributions with conjugate priors and Bayesian updating, which is what separates it from statrs and makes it the base of most Bayesian work in Rust. It is the anchor of a shelf you assemble rather than adopt — pair it with changepoint for online changepoints, ganesh for MCMC, friedrich or egobox-gp for Gaussian processes, adskalman for filtering. What does not exist anywhere in Rust is a probabilistic programming language you would build a project on: every attempt is a solo project under fifty stars.

Random variables

productionGrafana's augurs builds its time-series outlier detection on rv

version: 0.20.1
downloads: 257k / 90d
stars: 37
license: MIT
last commit: 2026-05-21
published: 2026-05-21
dependents: 6 owners · 7 repos

friedrich [crate · repo · docs]

Gaussian Process Regression.

version: 0.6.0
downloads: 1k / 90d
stars: 62
license: Apache-2.0
last commit: 2026-03-21
published: 2026-03-21
dependents: 3 owners · 1 repo

egobox-ego [crate · repo · docs]

A library for efficient global optimization

version: 0.40.1
downloads: 1k / 90d
stars: 185
license: Apache-2.0
last commit: 2026-09-11
published: 2026-08-31
dependents: 3 owners · 1 repo

egobox-gp [crate · repo · docs]

A library for gaussian process modeling

version: 0.36.3
downloads: 11k / 90d
stars: 185
license: Apache-2.0
last commit: 2026-09-11
published: 2026-08-31
dependents: 1 owner · 1 repo

changepoint [crate · repo · docs]

Change Point Detection in Rust

version: 0.15.0
downloads: 48k / 90d
stars: 36
license: MIT
last commit: 2025-10-20
published: 2025-10-20
dependents: 2 owners

lace [crate · repo · docs]

A probabilistic cross-categorization engine

version: 0.9.1
downloads: 113 / 90d
stars: 166
license: MIT
last commit: 2026-03-06
published: 2026-03-06
dependents: 1 owner

polya-gamma [crate · docs]

Efficient sampler for Polya-Gamma random variates

version: 0.5.3
downloads: 19k / 90d
license: MIT OR Apache-2.0
published: 2025-05-22
dependents: 1 owner

New & experimental (6)

Collected, but not clearing the published listing bar — too new, too quiet, or with no use by anyone but its author that we can see. Plenty of good software is here, some of it years old: this says what we can measure, not what it is worth.

adskalman [crate · repo · docs]

Kalman filter and Rauch-Tung-Striebel smoothing implementation using nalgebra, no_std

version: 0.18.0
downloads: 88k / 90d
stars: 91
license: MIT OR Apache-2.0
last commit: 2026-08-07
published: 2026-08-07
dependents: 1 owner · 2 repos

minikalman [crate · repo · docs]

A microcontroller targeted Kalman filter implementation

version: 0.7.0
downloads: 10k / 90d
stars: 21
license: MIT
last commit: 2026-04-25
published: 2026-04-25
dependents: 1 owner

ganesh [crate · repo · docs]

Minimization and sampling in Rust, simplified

version: 0.28.1
downloads: 1k / 90d
stars: 20
license: MIT OR Apache-2.0
last commit: 2026-07-16
published: 2026-07-16
dependents: 1 owner

ferric [crate · repo · docs]

A Probablistic Programming Language with a declarative syntax for random variables.

version: 0.2.1
downloads: 52 / 90d
stars: 42
license: MIT OR Apache-2.0
last commit: 2026-05-10
published: 2026-05-10

modppl [crate · repo · docs]

a experimental library for probabilistic programming in Rust.

version: 0.3.1
downloads: 49 / 90d
stars: 15
license: MIT
last commit: 2026-07-04
published: 2026-07-04

kalmanfilt [crate · repo · docs]

Kalman and other filters from Kalman-and-Bayesian-Filters-in-Python

version: 0.3.0
downloads: 53k / 90d
stars: 4
license: MIT
last commit: 2024-11-16
published: 2024-11-16
dependents: 1 owner

Time series & anomaly detection partial

Forecasting, decomposition and anomaly detection. Essentially one broad project, with a scattering of single-algorithm crates around it.

Featured: augurs [crate · repo · docs]

Time series in Rust is essentially this one project, and it is a good one: automatic exponential smoothing, MSTL decomposition, a real port of Prophet, dynamic time warping, seasonality detection, clustering, outlier detection and changepoints, across ten crates with Python and WASM bindings. It comes out of Grafana Labs, though its README notes it is not an official Grafana project. What it does not give you is a statsforecast-style model zoo or any deep forecasting model; outside augurs the shelf thins to dormant technical-indicator crates.

A time-series toolkit for forecasting, outlier detection, clustering and more.

talkaugurs, This Week in Rust Crate of the Week (2024-12-04) docsaugurs API documentation

version: 0.10.2
downloads: 18k / 90d
stars: 576
license: MIT OR Apache-2.0
last commit: 2026-08-25
published: 2026-02-24
dependents: 3 owners

changepoint [crate · repo · docs]

Change Point Detection in Rust

version: 0.15.0
downloads: 48k / 90d
stars: 36
license: MIT
last commit: 2025-10-20
published: 2025-10-20
dependents: 2 owners

deep_causality [crate · repo · docs]

Computational causality library. Provides causality graph, collections, context and causal reasoning.

version: 0.16.0
downloads: 28k / 90d
stars: 270
license: MIT
last commit: 2026-09-13
published: 2026-09-08
dependents: 1 owner

New & experimental (4)

Collected, but not clearing the published listing bar — too new, too quiet, or with no use by anyone but its author that we can see. Plenty of good software is here, some of it years old: this says what we can measure, not what it is worth.

anomaly_detection [crate · repo · docs]

Time series anomaly detection for Rust

version: 0.4.0
downloads: 795 / 90d
stars: 20
license: GPL-3.0-or-later
last commit: 2026-04-07
published: 2026-04-07
dependents: 2 owners

yata [crate · repo · docs]

Yet another Technical Analysis library. For rust now.

version: 0.7.0
downloads: 14k / 90d
stars: 399
license: Apache-2.0
last commit: 2024-09-19
published: 2024-03-07
dependents: 1 owner · 1 repo

krcf [crate · repo · docs]

Random Cut Forest implementation in Rust

version: 0.4.0
downloads: 14k / 90d
stars: 3
license: Apache-2.0
last commit: 2026-04-07
published: 2025-12-08

motif-rs [crate · repo · docs]

A modular matrix profile library for time series analysis

version: 0.1.0
downloads: 42 / 90d
license: MIT
last commit: 2026-02-20
published: 2026-02-20

Archive (3)

Kept so you stop looking for them. Archives are computed on every build and reverse themselves if the reason clears. Where a human has named the reason or a replacement, it cites the public issue behind that.

ta [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Technical analysis library. Implements number of indicators: EMA, SMA, RSI, MACD, Stochastic, etc.

version: 0.5.0
downloads: 70k / 90d
stars: 874
license: MIT
last commit: 2022-10-21
published: 2021-06-26
dependents: 14 owners · 9 repos

extended-isolation-forest [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

rust port of the anomaly detection algorithm

version: 0.2.3
downloads: 37k / 90d
stars: 18
license: MIT
last commit: 2023-04-26
published: 2022-11-30
dependents: 2 owners

arima [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

ARIMA time series modeling for Rust.

version: 0.3.0
downloads: 12k / 90d
stars: 40
license: non-standard
last commit: 2023-11-22
published: 2023-11-22
dependents: 1 owner

Optimization comprehensive

Linear and mixed-integer programming, convex conic solvers, and derivative-free and gradient-based optimizers. Rust is load-bearing outside Rust here: the standard Python convex-optimization package made a Rust solver its default.Source ¹: CVXPY

CVXPY, 2023-09-24 — CVXPY's maintainers, on their own default solver

Featured: good_lp [crate · repo · docs]

An ergonomic linear- and mixed-integer-programming API that dispatches to whichever solver you have: CBC, HiGHS, SCIP, Clarabel or the pure-Rust minilp. You write the model once in Rust expressions and change the backend with a Cargo feature, which is the thing that was missing when every solver had its own bindings and its own matrix layout. If you want the algebraic modeling layer above it — JuMP or Pyomo shaped — that arrived in 2026 as oximo.

Mixed Integer Linear Programming for Rust, with an user-friendly API. This crate allows modeling LP problems, and lets you solve them with various solvers.

benchmarkNaev benchmarks candidate LP solvers through good_lp, "since that is the API a replacement would use" productionCOIN-OR Cbc lists good_lp as its Rust interface

version: 1.15.3
downloads: 927k / 90d
stars: 458
license: MIT
last commit: 2026-08-31
published: 2026-08-06
dependents: 32 owners · 37 repos

Featured: clarabel [crate · repo · docs]

A pure-Rust interior-point solver for convex conic programs, from Oxford Control. It is the piece that makes "Rust can do convex optimization" true without shelling out to a C++ solver, and it is one of the backends good_lp can dispatch to. Together with good_lp, HiGHS and argmin this is the strongest ML-adjacent corner of the Rust numerical stack, and it is the one this site marked red for a decade.

Clarabel Conic Interior Point Solver for Rust / Python

productionCVXPY 1.5 switches its default solver from ECOS to Clarabel, and 1.6 removes ECOS as a dependency postThe CVXPY discussion proposing the ECOS-to-Clarabel move (2023-07)

version: 0.11.1
downloads: 894k / 90d
stars: 602
license: Apache-2.0
last commit: 2026-04-13
published: 2025-06-11
dependents: 22 owners · 1 repo

argmin [crate · repo · docs]

Numerical optimization in pure Rust

version: 0.11.0
downloads: 1.1M / 90d
stars: 1.3k
license: MIT OR Apache-2.0
last commit: 2025-10-02
published: 2025-09-28
dependents: 34 owners · 54 repos

highs [crate · repo · docs]

Safe rust bindings for the HiGHS linear programming solver. See http://highs.dev.

version: 2.4.0
downloads: 646k / 90d
stars: 42
license: MIT
last commit: 2026-07-09
published: 2026-07-09
dependents: 16 owners · 2 repos

coin_cbc [crate · repo · docs]

Rust bindings to the CoinOR CBC MILP Solveur using the C API.

version: 0.1.9
downloads: 27k / 90d
stars: 19
license: MIT
last commit: 2026-08-06
published: 2026-08-06
dependents: 6 owners · 23 repos

finitediff [crate · repo · docs]

Finite/numerical differentiation

version: 0.2.0
downloads: 49k / 90d
stars: 1.3k
license: MIT OR Apache-2.0
last commit: 2025-10-02
published: 2025-09-28
dependents: 11 owners · 6 repos

genetic_algorithm [crate · repo · docs]

A genetic algorithm implementation

version: 0.27.2
downloads: 2k / 90d
stars: 44
license: MIT OR Apache-2.0
last commit: 2026-06-04
published: 2026-06-04
dependents: 4 owners · 1 repo

russcip [crate · repo · docs]

Rust interface for SCIP

version: 0.10.0
downloads: 18k / 90d
stars: 97
license: Apache-2.0
last commit: 2026-09-04
published: 2026-06-24
dependents: 4 owners · 1 repo

egobox-ego [crate · repo · docs]

A library for efficient global optimization

version: 0.40.1
downloads: 1k / 90d
stars: 185
license: Apache-2.0
last commit: 2026-09-11
published: 2026-08-31
dependents: 3 owners · 1 repo

radiate [crate · repo · docs]

A Rust library for genetic algorithms and artificial evolution.

version: 1.3.1
downloads: 5k / 90d
stars: 256
license: MIT
last commit: 2026-09-13
published: 2026-09-13
dependents: 2 owners · 1 repo

cobyla [crate · repo · docs]

COBYLA optimizer for Rust

version: 1.0.4
downloads: 67k / 90d
stars: 28
license: non-standard
last commit: 2026-09-07
published: 2026-09-07
dependents: 2 owners · 2 repos

slsqp [crate · repo · docs]

SLSQP optimizer for Rust

version: 1.0.2
downloads: 9k / 90d
stars: 15
license: non-standard
last commit: 2026-08-24
published: 2026-08-24
dependents: 3 owners

lbfgs [crate · repo · docs]

A Rust implementation of the L-BFGS algorithm

version: 0.3.0
downloads: 113k / 90d
stars: 7
license: MIT OR Apache-2.0
last commit: 2026-03-11
published: 2025-11-13
dependents: 3 owners · 8 repos

optimization_engine [crate · repo · docs]

A pure Rust framework for embedded nonconvex optimization. Ideal for robotics!

version: 0.12.0
downloads: 107k / 90d
stars: 648
license: MIT OR Apache-2.0
last commit: 2026-03-31
published: 2026-03-31
dependents: 1 owner · 7 repos

globalsearch [crate · repo · docs]

A multistart framework for global optimization with scatter search and local NLP solvers written in Rust

version: 0.6.0
downloads: 1k / 90d
stars: 31
license: MIT
last commit: 2026-09-12
published: 2026-05-30

oximo [crate · repo · docs]

A Rust algebraic modeling library for solving optimization problems

version: 0.6.0
downloads: 151 / 90d
stars: 98
license: MIT OR Apache-2.0
last commit: 2026-09-13
published: 2026-08-23

New & experimental (7)

Collected, but not clearing the published listing bar — too new, too quiet, or with no use by anyone but its author that we can see. Plenty of good software is here, some of it years old: this says what we can measure, not what it is worth.

osqp [crate · repo · docs]

The OSQP (Operator Splitting Quadratic Program) solver.

version: 1.0.1
downloads: 94k / 90d
stars: 43
license: Apache-2.0
last commit: 2025-04-21
published: 2025-04-21
dependents: 7 owners · 4 repos

nlopt [crate · repo · docs]

Wrapper for the nlopt library

version: 0.8.1
downloads: 15k / 90d
stars: 33
license: MIT
last commit: 2025-03-26
published: 2025-03-26
dependents: 6 owners · 5 repos

ganesh [crate · repo · docs]

Minimization and sampling in Rust, simplified

version: 0.28.1
downloads: 1k / 90d
stars: 20
license: MIT OR Apache-2.0
last commit: 2026-07-16
published: 2026-07-16
dependents: 1 owner

simplers_optimization [crate · repo · docs]

A Rust implementation of the Simple(x) black-box optimization algorithm.

version: 0.5.0
downloads: 739 / 90d
stars: 33
license: Apache-2.0
last commit: 2026-05-12
published: 2026-05-12
dependents: 1 repo

cmaes [crate · repo · docs]

An implementation of the CMA-ES optimization algorithm.

version: 0.2.2
downloads: 6k / 90d
stars: 40
license: MIT OR Apache-2.0
last commit: 2025-05-19
published: 2024-12-12
dependents: 2 owners · 2 repos

mahf [crate · repo · docs]

A framework for modular construction and evaluation of metaheuristics.

version: 0.1.0
downloads: 120 / 90d
stars: 14
license: GPL-3.0-or-later
last commit: 2025-05-26
published: 2023-07-14
dependents: 2 owners

optirustic [crate · repo · docs]

NSGA-II and NSGA-III multi-objective optimization. The crates.io record is a tombstone — its own description says the crate moved — so the repository here points at where the work went.

version: 1.2.3
downloads: 156 / 90d
stars: 19
license: MIT
last commit: 2026-07-18
published: 2026-06-21

Archive (10)

Kept so you stop looking for them. Archives are computed on every build and reverse themselves if the reason clears. Where a human has named the reason or a replacement, it cites the public issue behind that.

pengowen123/eant2 [repo]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

EANT2 made in Rust.

stars: 15
license: GPL-3.0
last commit: 2022-06-12

minilp [crate · repo · docs]

Archived: repo archived. Detected automatically, and it un-archives itself if that changes.

A fast linear programming solver library.

version: 0.2.2
downloads: 72k / 90d
stars: 93
license: Apache-2.0
last commit: 2020-11-10
published: 2020-05-25
dependents: 5 owners · 43 repos

metaheuristics-nature [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

A collection of nature-inspired metaheuristic algorithms.

version: 10.1.0
downloads: 9k / 90d
stars: 16
license: MIT
last commit: 2024-05-02
published: 2024-05-02
dependents: 2 owners · 1 repo

metaheuristics [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Find approximate solutions to your optimisation problem using metaheuristics algorithms

version: 1.1.22
downloads: 1k / 90d
stars: 6
license: GPL-3.0
last commit: 2022-07-16
published: 2022-07-16
dependents: 1 owner · 2 repos

genevo [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

genevo provides building blocks to run simulations of optimization and search problems using genetic algorithms (GA). Execute genetic algorithm (GA) simulations in a customizable and extensible way.

version: 0.7.1
downloads: 3k / 90d
stars: 185
license: MIT/Apache-2.0
last commit: 2022-03-13
published: 2022-03-13
dependents: 1 owner · 6 repos

oxigen [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Fast, parallel, extensible and adaptable genetic algorithm library.

version: 2.2.2
downloads: 363 / 90d
stars: 185
license: MPL-2.0
last commit: 2021-02-28
published: 2021-02-28
dependents: 1 owner · 1 repo

darwin-rs [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Evolutionary algorithms library written in Rust.

version: 0.4.0
downloads: 87 / 90d
stars: 129
license: MIT
last commit: 2022-07-11
published: 2017-06-26
dependents: 1 repo

rsgenetic [crate · repo · docs]

Archived: repo archived. Detected automatically, and it un-archives itself if that changes.

A library providing genetic algorithm execution.

version: 1.8.1
downloads: 395 / 90d
stars: 78
license: MIT/Apache-2.0
last commit: 2021-01-22
published: 2021-01-21
dependents: 1 repo

evo [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Evolutionary Algorithm Library for Rust

version: 0.0.1
downloads: 7 / 90d
stars: 6
license: MIT
last commit: 2016-01-19
published: 2015-12-11

revonet [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Rust implementation of real-coded genetic algorithm for solving optimization problems and training of neural networks. The latter is also known as neuroevolution.

version: 0.2.1
downloads: 52 / 90d
stars: 20
license: GPL-3.0
last commit: 2017-08-20
published: 2017-08-20

Reinforcement learning too early

Agents, environments and algorithm implementations for reinforcement learning — a short shelf of mostly single-author projects, with no settled starting point on it.

Equim-chan/Mortal [repo]

🚀🀄️ A fast and strong AI for riichi mahjong, powered by Rust and deep reinforcement learning.

stars: 1.6k
license: AGPL-3.0
last commit: 2025-09-28

burn-rl [crate · repo · docs]

RL crate for the Burn framework

version: 0.22.0-pre.3
downloads: 41k / 90d
stars: 15.9k
license: MIT OR Apache-2.0
last commit: 2026-09-11
published: 2026-08-25
dependents: 3 owners

cervo [crate · repo · docs]

High level APIs for RL in games.

version: 0.9.2
downloads: 375 / 90d
stars: 56
license: MIT OR Apache-2.0
last commit: 2025-09-16
published: 2025-09-16
dependents: 1 owner · 1 repo

New & experimental (3)

Collected, but not clearing the published listing bar — too new, too quiet, or with no use by anyone but its author that we can see. Plenty of good software is here, some of it years old: this says what we can measure, not what it is worth.

bevy_rl [crate · repo · docs]

Build environments for reinforcement learning with bevy

version: 0.19.0-rc1
downloads: 292 / 90d
stars: 104
license: MIT OR Apache-2.0
last commit: 2026-06-22
published: 2026-06-22
dependents: 1 owner

r2l-core [crate · repo · docs]

A rust reinforcement learning library

version: 0.0.3
downloads: 382 / 90d
stars: 4
license: MIT
last commit: 2026-09-05
published: 2026-09-05
dependents: 1 owner

border [crate · repo · docs]

Reinforcement learning library

version: 0.0.8
downloads: 192 / 90d
stars: 67
license: GPL-2.0-or-later
last commit: 2025-05-17
published: 2025-05-17

Archive (4)

Kept so you stop looking for them. Archives are computed on every build and reverse themselves if the reason clears. Where a human has named the reason or a replacement, it cites the public issue behind that.

rurel [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Flexible, reusable reinforcement learning (Q learning) implementation

version: 0.6.0
downloads: 491 / 90d
stars: 165
license: MPL-2.0
last commit: 2024-06-25
published: 2024-06-25
dependents: 1 repo

NivenT/REnforce [repo]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

Reinforcement learning library written in Rust

stars: 24
last commit: 2017-08-03

benbaarber/rl [repo]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

A rust reinforcement learning library

stars: 51
last commit: 2024-08-03

rsrl [crate · repo · docs]

Archived: dormant. Detected automatically, and it un-archives itself if that changes.

A fast, extensible reinforcement learning framework in Rust

version: 0.8.1
downloads: 280 / 90d
stars: 206
license: MIT
last commit: 2020-06-18
published: 2020-06-18
dependents: 2 repos