-
cudarc
Safe and minimal CUDA bindings
-
stringzilla
Search, hash, sort, fingerprint, and fuzzy-match strings faster via SWAR, SIMD, and GPGPU
-
trueno
High-performance SIMD compute library with GPU support for matrix operations
-
trueno-zram-core
SIMD-accelerated LZ4/ZSTD compression for memory pages
-
iron_learn
ML library with GPU-accelerated gradient descent. Supports tensors, complex numbers, linear/logistic regression, and CUDA optimization.
-
realizar
Pure Rust ML inference engine built from scratch - model serving for GGUF and safetensors
-
dynamics
Molecular dynamics
-
rustacuda
CUDA Driver API Wrapper
-
gllm-kernels
Low-level attention kernels for gllm with CUDA/ROCm support
-
find_cuda_helper
Helper crate for searching for CUDA libraries
-
ringkernel-cuda
CUDA backend for RingKernel - NVIDIA GPU support via cudarc
-
fil-rustacuda
CUDA Driver API Wrapper
-
ringkernel-ecosystem
Ecosystem integrations for RingKernel - actors, web frameworks, data processing, ML
-
bindgen_cuda
Bindgen like interface to build cuda kernels to interact with within Rust
-
perdix
High-performance GPU-accelerated ring buffer for AI terminal multiplexing
-
cust
High level bindings to the CUDA Driver API
-
raw_preview_rs
processing raw images to create fast previews and handling EXIF metadata
-
cubecl-cuda
CUDA runtime for CubeCL
-
cuda-rust-wasm
CUDA to Rust transpiler with WebGPU/WASM support
-
cust_raw
Low level bindings to the CUDA Driver API
-
trueno-gpu
Pure Rust PTX generation for NVIDIA CUDA - no LLVM, no nvcc
-
cuda_setup
Assists with CUDA setup when using the CUDARC lib
-
trueno-explain
PTX/SIMD/wgpu visualization and tracing CLI for Trueno
-
uu_ptx
ptx ~ (uutils) display a permuted index of input
-
fatbinary
manipulate CUDA fatbinary format
-
slang-hal-build
Build script utilities for compile-time Slang shader compilation
-
ptx-90-parser
Parse NVIDIA PTX 9.0 assembly into a structured AST and explore modules via a CLI
-
sparkl3d-kernels
3-dimensional physics engine in Rust
-
vector-ta
High-performance technical analysis indicators with optional SIMD/CUDA and language bindings
-
luminal_cudarc
Safe wrappers around CUDA apis
-
torsh-jit
JIT compilation and kernel fusion for ToRSh deep learning framework
-
cuda_min
A minimal Rust lib for executing PTX code with nvidia GPU
-
cuda_builder
Builder for easily building rustc_codegen_nvvm crates
-
cuda_std
Standard library for CUDA with rustc_codegen_nvvm
-
nvvm
High level bindings to libnvvm
-
ptx_compiler
High level bindings to CUDA's ptx compilation APIs
-
scir-gpu
SciR GPU foundations: device arrays and CUDA (feature-gated) elementwise/FIR kernels with CPU parity
-
mistralrs_cudarc_fork
Safe wrappers around CUDA apis
-
ptx-linker
NVPTX modules linker
-
ringkernel-codegen
Code generation for RingKernel - generates GPU kernel source code
-
rustc_codegen_nvvm
A codegen backend for Rustc which targets the libnvvm CUDA library
-
mb2-wukong-expansion
Board crate for Elecfreaks Wukong Extension for the BBC micro:bit v2
-
ptx-parser
parser for ptx files
-
ptx-builder
NVPTX build helper
-
accel
GPGPU Framework for Rust
-
diffusionx-cuda-kernel
cuda kernels for diffusionx
-
cudarse-video
Bindings to the CUVID/NVDEC libraries
-
matrix_operations_cuda
perform matrix operations using cuda
-
cuda-oxide
high-level, rusty wrapper over CUDA. It provides the best safety one can get when working with hardware.
-
esb-ng
Nordic's Enhanced ShockBurst communication protocol
-
ptx-ir
Parallel Thread Execution (PTX) IR structure and parser
-
rustc-llvm-proxy
Dynamically proxy LLVM calls into Rust own shared library
-
ptoxide
A virtual machine to execute CUDA PTX without a GPU
-
cust_core
Core library for cust that can be shared across CPU and GPU
-
esb
Nordic's Enhanced ShockBurst communication protocol
-
rustacuda_core
Minimal kernel-support crate for Rustacuda
-
ssimulacra2-cuda-kernel
Ssimulacra2 CUDA implementation (device code)
-
del-fem-cudarc-kernel
Cuda PTX for the fem operations
-
ptx-90-parser-span-derive
Derive the Spanned helper trait used throughout ptx-90-parser AST types
-
ptx-90-parser-construct
Constructor helper macros used by the PTX 9.0 parser
-
wavesplatform
work with Waves blockchain (https://waves.tech/)
-
cust_derive
Macros for cust
-
nvptx
PTX compiler using LLVM/NVPTX target
-
accel-core
Support crate for writing GPGPU kernels using accel
-
accel-derive
Procedual macro for writing GPGPU kernel
Try searching with DuckDuckGo or on crates.io.