Fast inverse square root algorithm implementation.
Pre-defined constants from all disciplines (math, physics, ...)
Types for handling floats with type checking at compile time.
Embedded-friendly math library featuring fast floating point approximations (with small code size) for common arithmetic operations, trigonometry, 2D/3D vector types, statistical analysis, and quaternions. Optimizes for performance and small code size at the cost of precision.
Implementation of basic math data types with high level frontend and low level backend
Core tensor and autograd engine for ferrotorch — PyTorch in Rust
Forward-mode automatic differentiation with const-generic, stack-allocated first-, second-, and third-order jets (Jet1/Jet2/Jet3) for gradients, Hessians, and third-order tensors
A pure Rust, high-performance computer vision library focused on safety and portability.
A from-scratch deep learning library in Rust — tensors, autograd, NEON SIMD, Metal GPU, and working models
Zero-dependency 256-bit unsigned integer arithmetic for cryptographic applications
Pure-Rust JPEG XL (JXL) codec — RETIRED 2026-05-08 pending strict-isolation cleanroom workspace; current state ships round-1..3 ANS + headers + LfGlobal + GlobalModular wiring only. Rounds 7-11 + encoder rounds 1-6 retired with the libjxl-trace-reverse-engineering.md docs retirement of 2026-05-06.
Zero-dependency secp256k1 elliptic curve library for Bitcoin applications