Train models

Next-generation ML framework in pure Rust — `cargo install aprender` for the `apr` CLI

Ergonomic auto differentiation in Rust, with pytorch like apis.

ML framework on numr for training and inference on CPU, CUDA and WebGPU: attention, quantization, inference scheduling and distributed execution. Models include LLaMA, Mamba2, Qwen3.5, Whisper, Kokoro, VITS, VoxCPM2, Silero VAD, CLAP and super-resolution. Loads SafeTensors and GGUF.

Distributed runtime that replaces MPI between nodes: point-to-point messages, 11 collectives and RPC over QUIC, with no mpirun or MPI library.

Postgres with GPUs for ML/AI apps.

Python bindings

Bindings for the XLA C++ library.

Write and read tensorboard files.

Python bindings

An implementation of the RWKV language model in pure WebGPU.

Modern Rust bindings for Apple's Metal Performance Shaders Graph framework

A pure Rust implementation of the Fast Artificial Neural Network (FANN) library

Python bindings

Flexible, reusable reinforcement learning (Q learning) implementation

Tensors and differentiable operations in Rust

Machine Learning Framework for Hackers

floDl — a flow-graph deep learning framework built on libtorch

The neural network library implemented in Rust

A neural network implementation with a focus on cache-efficiency and sequential performance.

A fast, extensible reinforcement learning framework in Rust

Open AI environments bindings for Rust

Machine learning models, neural networks, metrics, and data utilities in Rust

A native-Rust deep learning framework with explicit memory safety and hardware acceleration

An Experimental Deep Learning Library

Build environments for reinforcement learning with bevy

Trains GPT-style language models from a YAML config and a tokenized dataset: dense GQA transformers, Mamba 1, 2 and 3, MLA and MoE layers on CPU or CUDA. Runs the training loop, optimizer, checkpoints, evaluation and metrics on boostr, and builds speech corpora through voxr.

A machine learning library for Rust.

ML library for directed acyclic tensor graphs.

A complete LSTM neural network library with training capabilities, multiple optimizers, and peephole variants.

Machine Learning Framework for Hackers

High-performance LLM fine-tuning framework for Apple Silicon

Python bindings

Wrapper for the Fast Artificial Neural Networks library

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

Tensors and dynamic neural networks.

Rust (candle) code for Build a LLM From Scratch by Sebastian Raschka

HAL: a machine learning library that is able to run on Nvidia, OpenCL or CPU BLAS based compute backends. It currently provides stackable classical neural networks, RNN's and soon to be LSTM's.

A differentiable computation graph for neural networks.

Wrapper around Microsoft CNTK library

Basic neural network library for classification.

An Open-Source Machine Learning Framework in Rust Δ

A cross-platform browser ML framework.

Reinforcement learning library written in Rust

An transformer based LLM. Written completely in Rust

A neural network crate

A learning-focused, high-performance tensor computation library built from scratch in Rust, featuring automatic differentiation and CPU/CUDA backends.

Deep Learning in Rust with the PyTorch API