ML Rust catalogue
Menu

All jobs

Tensors and GPU compute

Run tensor math with automatic gradients on your CPU or GPU.

19 projects. Checked .

Start here

The Featured projects with the most downloads.

All projects

Order: strongest tier first, then most downloads, then name, with unmeasured values last. A library counts downloads over 30 days. An app counts binary downloads per release. An app with none counts its crate downloads over 30 days.

What do the tiers mean?
Featured
Shown first. It passes every Featured gate.
Listed
Shown in the normal listing. No other tier applies.
New
A project whose first release is recent.
Archived
No longer maintained, or ended by its author.

Tiers come from published rules over raw signals. Stars alone never qualify a project.

Active dependents: crates from other owners that depend on this project in a current release. One owner counts at most once.

Movement needs more history.

ProjectDownloadsDependentsStarsLatest releaseRuns on (GPU)Native codeLicense
candle Featured

Minimalist ML framework.

Python bindings

 692 21,287 0.11.1 CUDA, Metal Apache-2.0 OR MIT
cubecl Featured

Multi-platform high-performance compute language extension for Rust.

 145 2,490 0.11.3 CUDA, WebGPU, Metal, ROCm, Vulkan Apache-2.0 OR MIT
burn Featured

Flexible and Comprehensive Deep Learning Framework in Rust

 391 16,257 0.22.1 CUDA, WebGPU, Metal, ROCm, Vulkan Apache-2.0 OR MIT
rten Featured

Tensor library for the RTen machine learning runtime

 29 360 0.27.0 Apache-2.0 OR MIT
Listed 15 projects
ug Listed Dormant

Micro compiler for tensor operations.

Python bindings

 3 75 0.5.0 CUDA, Metal Apache-2.0 OR MIT
dlpark Listed

Pure Rust DLPack implementation for zero-copy tensor interoperability

Python bindings

 5 66 0.10.0 CUDA Apache-2.0
dfdx Listed Dormant

Ergonomic auto differentiation in Rust, with pytorch like apis.

 4 1,944 0.13.0 CUDA Apache-2.0 OR MIT
numr Listed

The base for numerical code in Rust, in the role NumPy has in Python: n-dimensional tensors, linear algebra, FFT, statistics and autograd on CPU, CUDA and WebGPU. The backends use native kernels, with no cuBLAS, MKL or other vendor math library.

 7 59 0.15.2 CUDA, WebGPU Apache-2.0
boostr Listed

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.

 7 20 0.11.2 CUDA, WebGPU Apache-2.0
rust-autograd Listed

Tensors and differentiable operations in Rust

 0 504 1.1.1 MIT
oxicuda Listed

OxiCUDA replaces the entire NVIDIA CUDA Toolkit software stack with type-safe, memory-safe Rust code.

36 145 0.5.5 WebGPU, Metal, ROCm, Vulkan Apache-2.0
emu Listed Dormant

A low-level GPGPU compute library featuring cross-platform support and SPIR-V as input

 0 1,604 0.1.1 WebGPU MIT
autograph Listed Dormant

A machine learning library for Rust.

 0 333 0.2.1 Apache-2.0 OR MIT
juice Listed Dormant

Machine Learning Framework for Hackers

 5 1,129 0.3.0 CUDA Apache-2.0 OR MIT
neuronika Listed Dormant

Tensors and dynamic neural networks.

 0 1,083 0.2.0 MPL-2.0
L2 App Listed Dormant

l2 is a fast, Pytorch-style Tensor+Autograd library written in Rust

 Not applicable 197 Not measured Not measured MIT
mushin Listed Dormant

Computational graphs with reverse automatic differentation in the GPU

 0 396 0.5.0 Apache-2.0 OR MIT
RustTensor App Listed Dormant

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

 Not measured Not applicable 79 Not measured Not measured Not measured MIT
wgmath App Listed

GPU scientific computing on every platform

 Not measured Not applicable 150 Not measured Not measured Not measured Apache-2.0