#deep-learning #pytorch #machine-learning

no-std burn-core

Flexible and Comprehensive Deep Learning Framework in Rust

37 releases

Uses new Rust 2024

new 0.21.0 May 7, 2026
0.21.0-pre.4 Apr 27, 2026
0.21.0-pre.2 Mar 2, 2026
0.20.0-pre.6 Dec 18, 2025
0.6.0 Mar 21, 2023

#68 in #pytorch

Download history 16001/week @ 2026-01-21 17710/week @ 2026-01-28 18481/week @ 2026-02-04 15434/week @ 2026-02-11 15829/week @ 2026-02-18 16089/week @ 2026-02-25 17773/week @ 2026-03-04 18493/week @ 2026-03-11 18095/week @ 2026-03-18 18291/week @ 2026-03-25 18157/week @ 2026-04-01 15006/week @ 2026-04-08 18842/week @ 2026-04-15 19212/week @ 2026-04-22 22134/week @ 2026-04-29 25605/week @ 2026-05-06

88,481 downloads per month
Used in 284 crates (12 directly)

MIT/Apache

1.5MB
26K SLoC

The core crate of Burn. Structs and macros used by most projects. Add use burn::prelude::* to your code to quickly get started with Burn.


Burn Core

This crate should be used with burn. It contains the core traits and components for building and training deep learning models with Burn.

Current Crates.io Version license

Feature Flags

This crate can be used without the standard library (#![no_std]) with alloc by disabling the default std feature.

  • std - enables the standard library. Enabled by default.

Dependencies

~14–67MB
~1M SLoC