-
ort
A safe Rust wrapper for ONNX Runtime 1.23 - Optimize and accelerate machine learning inference & training
-
hf-hub
This crates aims ease the interaction with huggingface It aims to be compatible with the huggingface_hub python package, but only implements a smaller subset of functions
-
candle-nn
Minimalist ML framework
-
burn
Flexible and Comprehensive Deep Learning Framework in Rust
-
candle-transformers
Minimalist ML framework
-
candle-core
Minimalist ML framework
-
lance
A columnar data format that is 100x faster than Parquet for random access
-
tch
Rust wrappers for the PyTorch C++ api (libtorch)
-
smartcore
Machine Learning in Rust
-
linfa
A Machine Learning framework for Rust
-
rv
Random variables
-
jams
CLI application for interaction with J.A.M.S - Just Another Model Server
-
rstats
Statistics, Information Measures, Data Analysis, Linear Algebra, Clifford Algebra, Machine Learning, Geometric Median, Matrix Decompositions, Mahalanobis Distance, Hulls, Multithreading
-
torch-sys
Low-level FFI bindings for the PyTorch C++ api (libtorch)
-
mlx-rs
Unofficial rust wrapper for Apple's mlx machine learning library
-
candle-onnx
ONNX support for Candle
-
lightgbm3
Rust bindings for LightGBM library
-
openvino
High-level bindings for OpenVINO
-
constriction
Entropy coders for research and production (Rust and Python)
-
openai
An unofficial Rust library for the OpenAI API
-
scirs2-text
Text processing module for SciRS2 (scirs2-text)
-
candle-metal-kernels
Metal kernels for Candle
-
perpetual
A self-generalizing gradient boosting machine that doesn't need hyperparameter optimization
-
oar-ocr
An Optical Character Recognition (OCR) and Document Layout Analysis library written in Rust
-
mistralrs
Fast, flexible LLM inference
-
rgwml
ONLY 🤯 RUST-dominant AI, Data Science & Machine Learning RUST Library designed to minimize developer cognitive load, and replicate the Python Pandas Library with OpenAI, XGBoost…
-
petal-clustering
A collection of clustering algorithms
-
fsrs
including Optimizer and Scheduler
-
scirs2-vision
Computer vision module for SciRS2 (scirs2-vision)
-
moshi
real-time voice AI
-
ocrs
OCR engine
-
candle-flash-attn
Flash attention layer for the candle ML framework
-
ruvector-attention
Attention mechanisms for ruvector - geometric, graph, and sparse attention
-
aprender
Next-generation machine learning library in pure Rust
-
scirs2-neural
Neural network building blocks module for SciRS2 (scirs2-neural) - Minimal Version
-
lance-bitpacking
Vendored copy of https://github.com/spiraldb/fastlanes for use in Lance
-
hdbscan
clustering in pure Rust. A huge improvement on DBSCAN, capable of identifying clusters of varying densities.
-
scirs2-metrics
Machine Learning evaluation metrics module for SciRS2 (scirs2-metrics)
-
scirs2-transform
Data transformation module for SciRS2 (scirs2-transform)
-
torsh-linalg
Linear algebra operations for ToRSh with SciRS2 integration
-
aws-sdk-codegurureviewer
AWS SDK for Amazon CodeGuru Reviewer
-
changepoint
Change Point Detection in Rust
-
aws-sdk-sagemakera2iruntime
AWS SDK for Amazon Augmented AI Runtime
-
changeforest
Random Forests for Change Point Detection
-
aws-sdk-codeguruprofiler
AWS SDK for Amazon CodeGuru Profiler
-
autogpt
🦀 A Pure Rust Framework For Building AGIs
-
aws-sdk-personalize
AWS SDK for Amazon Personalize
-
xgb
Machine learning using XGBoost
-
aws-sdk-lookoutequipment
AWS SDK for Amazon Lookout for Equipment
-
candle-kernels
CUDA kernels for Candle
-
birdnet-onnx
Bird species detection using BirdNET, Perch, and BSG Finland ONNX models
-
lgbm
Unofficial Rust bindings for LightGBM
-
easy-ml
Machine learning library providing matrices, named tensors, linear algebra and automatic differentiation aimed at being easy to use
-
sklears-metrics
Evaluation metrics for sklears: accuracy, precision, recall, F1, ROC-AUC, etc
-
langdb_clust
An unofficial Rust client for the Anthropic/Claude API
-
candle-examples
Minimalist ML framework
-
scirs2-datasets
Datasets module for SciRS2 (scirs2-datasets)
-
tensor-tools
Minimalist ML framework
-
rustyml
A high-performance machine learning & deep learning library in pure Rust, offering ML algorithms and neural network support
-
rustorch
Production-ready PyTorch-compatible deep learning library in Rust with special mathematical functions (gamma, Bessel, error functions), statistical distributions, Fourier transforms (FFT/RFFT)…
-
ngboost-rs
Natural Gradient Boosting for Probabilistic Prediction - A Rust implementation of NGBoost
-
float4
MXFP4-compatible 4-bit floating point types and block formats for Rust
-
kornia
low-level 3D computer vision library in Rust
-
entrenar
Training & Optimization library with autograd, LoRA, quantization, and model merging
-
torsh-nn
Neural network modules for ToRSh with PyTorch-compatible API
-
ug
Micro compiler for tensor operations
-
peft-rs
Comprehensive PEFT (Parameter-Efficient Fine-Tuning) adapter library for Rust
-
scirs2-graph
Graph processing module for SciRS2 (scirs2-graph)
-
executorch
Rust bindings for ExecuTorch - On-device AI across mobile, embedded and edge for PyTorch
-
pmetal-mlx-rs
pmetal-maintained fork of mlx-rs: unofficial rust wrapper for Apple's mlx machine learning library
-
aws-sdk-neptunedata
AWS SDK for Amazon NeptuneData
-
aphelion-core
Core library for Aphelion AI framework - unified frontend for AI model development
-
aws-sdk-codegurusecurity
AWS SDK for Amazon CodeGuru Security
-
bhtsne
Exact and Barnes-Hut implementations of t-SNE
-
candle-optimisers
Optimisers for use with candle, the minimalist ML framework
-
realizar
Pure Rust ML inference engine built from scratch - model serving for GGUF and safetensors
-
libsvm-rs
Pure Rust reimplementation of LIBSVM — SVM training and prediction
-
lance-datagen
A columnar data format that is 100x faster than Parquet for random access
-
tl-lang
A differentiable programming language with tensor support for machine learning
-
lance-datafusion
Internal utilities used by other lance modules to simplify working with datafusion
-
dynamo-runtime
Dynamo Runtime Library
-
fsst
FSST string compression for Lance
-
iron_learn
ML library with GPU-accelerated gradient descent. Supports tensors, complex numbers, linear/logistic regression, and CUDA optimization.
-
kmedoids
k-Medoids clustering with the FasterPAM algorithm
-
valentinus
A thread-safe vector database for model inference inside LMDB
-
pav_regression
The pair adjacent violators algorithm for isotonic regression
-
torsh-tensor
Tensor implementation for ToRSh with PyTorch-compatible API
-
nabla-ml
A numpy-like library for Rust
-
scirs2-autograd
Automatic differentiation module for SciRS2 (scirs2-autograd)
-
gguf-rs-lib
reading and writing GGUF (GGML Universal Format) files
-
coglet
High-performance prediction server for Cog ML models
-
darjeeling
Machine learning tools for rust. Currently comprising of categorization and generation neural network model training
-
is-it-slop
Detect AI-generated slop text using machine learning
-
ocr-rs
A lightweight and efficient OCR library based on PaddleOCR models, using the MNN inference framework for high-performance text detection and recognition
-
concision
toolkit for designing machine-learning models in Rust
-
neat
working with NEAT in rust
-
fast-umap
Configurable UMAP (Uniform Manifold Approximation and Projection) in Rust
-
rust-bert
Ready-to-use NLP pipelines and language models
-
jax-rs
JAX in Rust - A complete machine learning framework with WebGPU acceleration
-
arrowspace
Graph-based vector search with taumode (λτ) indexing
-
forust-ml
A lightweight gradient boosting implementation in Rust
-
metal-candle
Production-quality Rust ML crate for Apple Silicon - LoRA training, inference, and text generation using Candle with Metal backend
-
tensormq-broker
A zero-copy, high-performance tensor message broker
-
cake-cli
Distributed LLM inference for mobile, desktop and server
-
axonml-tensor
N-dimensional tensor operations for the Axonml ML framework
-
triplets
Composable data sampling primitives for deterministic multi-source ML/AI training-data orchestration
-
similari-trackers-rs
Machine learning framework for building object trackers and similarity search engines
-
ghostflow-nn
Neural network layers for GhostFlow ML framework
-
karma
A sophisticated Hidden Markov Model (HMM) implementation using the Baum-Welch algorithm
-
oxidizr
A Rust-based LLM training framework built on Candle
-
burn-ndarray
Ndarray backend for the Burn framework
-
flowly
modular and reusable components for building robust pipelines processing audio, video and other
-
burn-remote
Backend router decorator over the network
-
tflitec
A safe Rust wrapper of TensorFlow Lite C API supporting x86_64 and ARM (iOS, Android)
-
burn-vision
Vision processing operations for burn tensors
-
ruv-fann
A pure Rust implementation of the Fast Artificial Neural Network (FANN) library
-
ghostflow-ml
Classical ML algorithms for GhostFlow
-
lele
Bare-metal machine learning inference library in Rust
-
llms-from-scratch-rs
Rust (candle) code for Build a LLM From Scratch by Sebastian Raschka
-
eryon-mem
implements the memory-related aspects of the eryon framework
-
burn-store
Storage and serialization infrastructure for Burn
-
flowly-spsc
Flowly is a library of modular and reusable components for building robust pipelines processing audio, video and other
-
train-station
A high-performance, PyTorch inspired, zero dependency Rust machine learning library
-
qwen3_tts
Rust port of Qwen3 TTS - Text-to-Speech model
-
mistralrs-quant
Fast, flexible LLM inference
-
influence
CLI tool for downloading HuggingFace models and running local LLM inference
-
hextral
Comprehensive neural network library with dataset loading, batch normalization, 9 activation functions, 5 loss functions, multiple optimizers, regularization, and clean async-first API
-
halldyll_starter_runpod
managing RunPod GPU pods - Provisioning, orchestration & state management
-
ghostflow-cuda
CUDA backend for GhostFlow ML framework
-
rust-lstm
A complete LSTM neural network library with training capabilities, multiple optimizers, and peephole variants
-
mcp-sdk-rs
An SDK for building Model Context Protocol (MCP) clients and servers in Rust
-
flowly-io
Flowly is a library of modular and reusable components for building robust pipelines processing audio, video and other
-
burn-tch
LibTorch backend for the Burn framework using the tch bindings
-
runnt
Easy Neural Network for machine learning
-
kalman_clustering
Clustering vectors using Kalman Filter
-
rusty_scrfd
A high-performance face detection library using SCRFD model with OpenCV integration
-
oxide-rs
AI Inference library and CLI in Rust - llama.cpp style
-
hf-fetch-model
Fast HuggingFace model downloads for Rust — an embeddable library for downloading HuggingFace models with maximum throughput
-
nabled
A linear algebra library written in Rust with an ndarray-first architecture
-
ghostflow-core
Core tensor operations for GhostFlow ML framework - optimized for maximum performance
-
qwen_tts
Qwen3-TTS text-to-speech model implementation for Candle
-
single_algebra
A linear algebra convenience library for the single-rust library. Can be used externally as well.
-
temporal-compare
High-performance framework for benchmarking temporal prediction algorithms inspired by Time-R1
-
rustkernel-cli
CLI tool for RustKernels management
-
ort-sys
Unsafe Rust bindings for ONNX Runtime 1.23 - Optimize and Accelerate Machine Learning Inferencing
-
atlas-cli
Machine Learning Lifecycle & Transparency Manager - Create and verify manifests for ML models and datasets
-
burn-tensor
Tensor library with user-friendly APIs and automatic differentiation support
-
metatensor
Self-describing sparse tensor data format for atomistic machine learning and beyond
-
lance-namespace-impls
Lance Namespace Implementations
-
candle-datasets
Minimalist ML framework
-
rusty-logging
Logging for OpsML
-
rust-ai-core
Unified AI engineering toolkit: orchestrates peft-rs, qlora-rs, unsloth-rs, axolotl-rs, bitnet-quantize, trit-vsa, vsa-optim-rs, and tritter-accel
-
runnx
A minimal, verifiable ONNX runtime implementation in Rust
-
hybrid-predict-trainer-rs
Hybridized predictive training framework with warmup, full-train, predict, and residual correction phases for accelerated deep learning
-
nnl
A high-performance neural network library for Rust with CPU and GPU support
-
hodu
A user-friendly ML framework built in Rust for rapid prototyping and embedded deployment
-
cyanea-ml
ML primitives for bio (embeddings, etc.) for the Cyanea bioinformatics ecosystem
-
quantrs2-ml
Quantum Machine Learning module for QuantRS2
-
bitnet-training
Training and fine-tuning utilities for BitNet models
-
liboxen
Oxen is a fast, unstructured data version control, to help version datasets, written in Rust
-
dlpk
DLPack integration for Rust
-
ghostflow-autograd
Automatic differentiation for GhostFlow ML framework
-
relayrl_framework
A distributed, system-oriented multi-agent reinforcement learning framework
-
minimum_ml
Experimental Machine Learning Library
-
uni-xervo
Unified Rust runtime for local and remote embedding, reranking, and generation models
-
mistralrs-server-core
Fast, flexible LLM inference
-
petite-ad
A lightweight automatic differentiation library for learning and educational purposes
-
concision-traits
implements the core modules for the concision framework
-
candle-pyo3
Minimalist ML framework
-
treeboost
High-performance Gradient Boosted Decision Tree engine for large-scale tabular data
-
lance-file
Lance file format
-
wgc
An ergonomic Rust wrapper for Windows.Graphics.Capture API
-
lance-namespace
Lance Namespace Core APIs
-
lace_cc
Core of the Lace cross-categorization engine library
-
lance-encoding
Encoders and decoders for the Lance file format
-
mlmf
Machine Learning Model Files - Loading, saving, and dynamic mapping for ML models
-
lance-table
Lance table format
-
rust_metrics
Incremental evaluation metrics for various machine learning pipelines
-
egraph
variant machine learning
-
ghostflow-data
Data loading utilities for GhostFlow ML framework
-
onnx-extractor
Lightweight ONNX model parser for extracting tensor shapes, operations, and data
-
pkboost
Shannon-guided gradient boosting for extreme class imbalance with adaptive drift detection. Outperforms XGBoost/LightGBM on imbalanced data.
-
linfa-nn
A collection of nearest neighbour algorithms
-
tenflowers-ffi
Python bindings for TenfloweRS
-
lance-io
I/O utilities for Lance
-
lance-linalg
A columnar data format that is 100x faster than Parquet for random access
-
burn-wgpu
WGPU backend for the Burn framework
-
lance-index
Lance indices implementation
-
candle-flash-attn-v3
Flash attention v3 layer for the candle ML framework
-
papers-openalex
Rust client for the OpenAlex academic research API
-
burn-fusion
Kernel fusion backend decorator for the Burn framework
-
tensorrs
Tensors is a lightweight machine learning library in Rust
-
optimiz-rs
High-performance optimization algorithms in Rust with Python bindings
-
burn-train
Training crate for the Burn framework
-
embellama
High-performance Rust library for generating text embeddings using llama-cpp
-
burn-cubecl
Generic backend that can be compiled just-in-time to any shader language target
-
tsetlin-rs
A professional Rust implementation of the Tsetlin Machine algorithm
-
linfa-linear
A Machine Learning framework for Rust
-
tensor_optim
A tiny tensor engine lacking compromise on portability
-
lance-arrow
Arrow Extension for Lance
-
manopt-rs
A high-performance Rust library for manifold optimization built on the Burn deep learning framework
-
cervo
High level APIs for RL in games
-
hyperparameter
A high performance configuration system for Rust
-
burn-autodiff
Automatic differentiation backend for the Burn framework
-
voxudio
A real-time audio processing library with ONNX runtime support
-
tenflowers
Pure Rust implementation of TensorFlow - A comprehensive deep learning framework
-
automl
Automated machine learning for classification and regression
-
cake-core
Distributed LLM inference for mobile, desktop and server
-
libinfer
Rust interface to TensorRT for high-performance GPU inference
-
optirs
Advanced ML optimization and hardware acceleration library (main integration crate)
-
instmodel_inference
High-performance neural network inference library with instruction-based execution
-
mlx-rs-burn
Unofficial rust wrapper for Apple's mlx machine learning library (fork with slice, scatter, flip operations for burn-mlx)
-
airust
Trainable, modular AI engine in Rust with compile-time knowledge
-
pmetal
High-performance LLM fine-tuning framework for Apple Silicon
-
torsh
A blazingly fast, production-ready deep learning framework written in pure Rust
-
fastkmeans-rs
A fast and efficient k-means clustering implementation in Rust, compatible with ndarray
-
tensorlogic
Logic-as-Tensor planning layer - meta crate re-exporting all TensorLogic components
-
rhmm
hidden markov models
-
bevy_zeroverse
bevy zeroverse synthetic dataset generator
-
cervo-cli
High level APIs for RL in games
-
ryft
A Framework for Tracing, Automatic Differentiation, and Just-In-Time Compilation
-
linfa-reduction
A collection of dimensionality reduction techniques
-
vsa-optim-rs
Deterministic training optimization using VSA compression and closed-form gradient prediction
-
atlas-c2pa-lib
creating, signing, and verifying machine learning assets with C2PA
-
instmodel
Instruction-based neural network model training library using Burn
-
lance-geo
Lance's geospatial extension providing geospatial UDFs
-
ghostflow-optim
Optimizers for GhostFlow ML framework
-
torsh-special
Special mathematical functions for ToRSh with SciRS2 integration
-
ffsvm
A libSVM compatible support vector machine, but up to 10x faster, for games or VR
-
eval-metrics
Evaluation metrics for machine learning
-
mistralrs-cli
Command-line interface for mistral.rs LLM inference
-
tenflowers-autograd
Automatic differentiation engine for TenfloweRS
-
lance-testing
A columnar data format that is 100x faster than Parquet for random access
-
qora-fl
Qora: Quorum-Oriented Robust Aggregation for Federated Learning
-
conformal-prediction
Conformal prediction with formal verification: CPD, PCP, streaming calibration, and Lean4 proofs
-
bevy_burn
bevy burn bindings
-
lance-jni
JNI bindings for Lance Columnar format
-
hanzo-ug
Minimalist ML framework
-
catboost-rust
Rust bindings for CatBoost, a gradient boosting library for machine learning. Downloads CatBoost binaries at runtime for cross-platform compatibility.
-
egobox-ego
efficient global optimization
-
trustformers
port of Hugging Face Transformers
-
rustyasg
A deep learning framework in Rust, focusing on an Abstract Semantic Graph (ASG) for autograd and Transformer architectures
-
tenflowers-neural
Neural network layers, models and training APIs for TenfloweRS
-
lumen-core
A tiny ML framework
Try searching with DuckDuckGo.