Lib.rs

nvml-wrapper

A safe and ergonomic Rust wrapper for the NVIDIA Management Library

opencl3

Khronos OpenCL 3.0 API and extensions

lru-slab

Pre-allocated storage with constant-time LRU tracking

whisper-rs

Rust bindings for whisper.cpp

vulkano

Safe wrapper for the Vulkan graphics API

femtovg

Antialiased 2D vector drawing library

pixels

A tiny library providing a GPU-powered pixel frame buffer

ktx2

Parser for the ktx2 texture container format

surfman

A cross-platform, low-level toolkit for GPU surface management

cl3

Khronos OpenCL 3.0 API and extensions

neptune

Poseidon hashing over BLS12-381 for Filecoin

cubecl

Multi-platform high-performance compute language extension for Rust

spirv-std

Standard functions and types for SPIR-V

ocl

OpenCL bindings and interfaces for Rust

webrender

A GPU accelerated 2D renderer for web content

dynamo-runtime

Dynamo Runtime Library

qmassa

Terminal-based tool for displaying GPUs usage stats on Linux

pathfinder_simd

SIMD library

kubectl-view-allocations

kubectl plugin to list allocations (cpu, memory, gpu,...) X (utilization, requested, limit, allocatable,...)

gpu-descriptor

agnostic descriptor allocator for Vulkan like APIs

xilem

A next-generation cross-platform Rust UI framework

inlyne

Introducing Inlyne, a GPU powered yet browserless tool to help you quickly view markdown files in the blink of an eye

neotop

Linux-first terminal system monitor: per-core CPU spectrum with SMT/NUMA, NVIDIA/AMD/Intel GPU dashboards with i915 per-engine breakdown, KVM hypervisor insight, container/runtime process grouping, Catppuccin themes

poulpy-hal

providing layouts and a trait-based hardware acceleration layer with open extension points, matching the API and types of spqlios-arithmetic

cuvs

RAPIDS vector search library

mlx-native

Pure-Rust Metal GPU compute library for MLX-compatible inference on Apple Silicon

axonml-autograd

Automatic differentiation engine for Axonml ML framework

tract-gpu

Tiny, no-nonsense, self contained, TensorFlow and ONNX inference

precord

Command line tool for recording process or system performance data

tracy_full

Fully featured bindings for the Tracy profiler

ai-hwaccel

Universal AI hardware accelerator detection, capability querying, and workload planning for Rust

ringkernel-cuda

CUDA backend for RingKernel - NVIDIA GPU support via cudarc

system_info_collector

fast application to collect os information and create graphs based on it

gpu-alloc

agnostic memory allocator for Vulkan like APIs

diffusionx

A multi-threaded crate for random number generation and stochastic process simulation, with optional GPU acceleration

beamterm-renderer

High-performance WebGL2 terminal renderer for beamterm, targeting sub-millisecond render times in web browsers

spark-dashboard

Real-time hardware and LLM inference monitoring for Linux hosts with NVIDIA GPUs

resource-tracker

Lightweight Linux resource and GPU tracker for system and process monitoring

torc

Workflow management system

pathfinder_geometry

Basic SIMD-accelerated geometry/linear algebra

aprender

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

ctt

Compress images to GPU texture formats

prgpu

GPU-accelerated rendering utilities for Adobe Premiere Pro and After Effects plugins

warp-types

Type-safe GPU warp programming via linear typestate: compile-time prevention of shuffle-from-inactive-lane bugs

vhost-device-gpu

A virtio-gpu device using the vhost-user protocol

oxicuda-driver

OxiCUDA Driver - Dynamic CUDA driver API wrapper via libloading (zero SDK dependency)

mold-ai

Local AI image generation CLI — FLUX, SDXL, SD3.5, Z-Image diffusion models on your GPU

mold-ai-discord

Discord bot for mold — AI image generation via slash commands

skia-canvas

GPU-accelerated, multi-threaded HTML Canvas-compatible 2D rendering for Rust and Node, powered by Skia

nvml-wrapper-sys

Generated bindings to the NVIDIA Management Library

gpu-setfan

GPU fan control via temperature setpoint

rocm-rs

Rust bindings for AMD ROCm libraries

nv-swaptop

A terminal user interface tool to monitor swap, NUMA topology, and GPU memory usage

oxicuda-ptx

OxiCUDA PTX - PTX code generation DSL and IR for GPU kernel development

shiguredo_vpl

Rust bindings for Intel oneVPL (Video Processing Library)

cudaforge

Advanced CUDA kernel builder for Rust with incremental builds, auto-detection, and external dependency support

neomacs

Standalone Rust binary for Neomacs (no C dependency)

envy-tui

TUI manager for EnvyControl - GPU switching for Nvidia Optimus laptops

wgpu-llm-cli

Terminal-based chat interface for the wgpu LLM inference engine

apple-metal

Safe Rust bindings for Apple's Metal framework — devices, resources, command encoding, advanced GPU objects, and IOSurface interop on macOS, backed by a Swift bridge

av-denoise

Fast and efficient video denoising using accelerated nlmeans

tl_backend

GPU Backend Trait Definitions for TL

ranga

Core image processing library — color spaces, blend modes, pixel buffers, filters, and GPU compute for Rust

burn-onnx

importing ONNX models into the Burn framework

hypomnesis

External RAM and VRAM measurement for Rust processes (process RSS plus per-process and device-wide GPU memory: Windows DXGI + NVML, Linux NVML, with nvidia-smi fallback)

concerto-cli

Concerto CLI: a Rust inference multiplexer for self-hosted LLMs. Orchestrates vLLM, llama.cpp, and SGLang processes across 1-8 GPUs with demand-aware routing.

zust-vm-spirv

SPIR-V code generation backend for the Zust scripting language

mold-ai-server

HTTP inference server for mold

mtl-gpu

Rust bindings to Apple's Metal framework

ringkernel-wavesim

Interactive 2D wave propagation showcase for RingKernel

axonml-data

Data loading utilities for the Axonml ML framework

fust

A fast system info tool written in Rust

torsh-backend

Backend abstraction layer for ToRSh

cubecl-cpp

CPP transpiler for CubeCL

gpu-fft

performing Fast Fourier Transform (FFT) and Inverse FFT using GPU acceleration

mosec

Model Serving made Efficient in the Cloud

oxicuda-dnn

OxiCUDA DNN - GPU-accelerated deep learning primitives (cuDNN equivalent)

gpu-trace-perf

Plays a collection of GPU traces under different environments to evaluate driver changes on performance

scenix-renderer

wgpu renderer, GPU scene upload, passes, and frame orchestration for scenix

glcore-rs

The OpenGL core functions for Rust, also supports OpenGL ES

fast-umap

Configurable UMAP (Uniform Manifold Approximation and Projection) in Rust

metal-softfloat

IEEE-754 binary64 (f64) software emulation for Apple Metal compute kernels and pure Rust

runpod-cli

RunPod CLI — auto-generated from OpenAPI spec

quant-iron

high-performance, hardware-accelerated modular quantum computing library with a focus on physical applications. Quant-Iron provides tools to represent quantum states, apply standard quantum gates…

zust-vm-metal

Metal shader source generation backend for the Zust scripting language

ryft-pjrt

Ryft bindings for PJRT

ringkernel-ecosystem

Ecosystem integrations for RingKernel - actors, web frameworks, data processing, ML

dsfb-gpu-debug-cuda

CUDA FFI bridge and kernel dispatch for dsfb-gpu-debug. Builds without nvcc unless the cuda feature is set.

qjl-sketch

QJL sign-based vector compression and scoring with near-optimal distortion rate

kn-cuda-eval

A CUDA executor for neural network graphs

metaltile-std

MetalTile kernel standard library — benchmark metadata and type definitions

oversee

A modern system monitor for macOS with Apple Silicon GPU support

numr

High-performance numerical computing with multi-backend GPU acceleration (CPU/CUDA/WebGPU)

hive-gpu

High-performance GPU acceleration for vector operations with Device Info API (Metal, CUDA, ROCm)

dsfb-gpu-debug-demo

CLI binary for dsfb-gpu-debug: generate fixtures, run CPU/GPU pipelines, compare case files

beamterm-atlas

Font atlas generator for beamterm WebGL terminal renderer, creating GPU-optimized texture arrays from TTF/OTF fonts

scrying

Capability-driven system-webview frame adapter — scry into WebView2/WKWebView/WebKitGTK and produce wgpu-importable frames

oxicuda-memory

OxiCUDA Memory - Type-safe GPU memory management with Rust ownership semantics

dsfb-gpu-atlas-corpus

Literature detector corpus and canonicalisation court for DSFB-GPU-Atlas

burn-vision

Vision processing operations for burn tensors

runmat-thread-local

Cross-platform thread-local storage helpers for RunMat (native and wasm)

cudaclaw

CUDA Rust bindings for GPU programming in the Cocapn fleet

oxicuda

Pure Rust CUDA replacement for the COOLJAPAN ecosystem (95% performance target)

vkcore-rs

The Vulkan core functions for Rust

zeusd

Zeus daemon

vulkane

Vulkan API bindings generated entirely from vk.xml, with a complete safe RAII wrapper covering compute and graphics: instance/device/queue, buffer, image, sampler, render pass, framebuffer…

oxicuda-fft

OxiCUDA FFT - GPU-accelerated FFT operations (cuFFT equivalent)

oxicuda-launch

OxiCUDA Launch - Type-safe GPU kernel launch infrastructure

easy-async-opencl3

A declarative, multi-device asynchronous executor for OpenCL based on cl3

gloss-rs

Top level crate for gloss-rs

kn-runtime

Dynamic wrapper around CPU and GPU inference

narsil

A terminal-based system resource monitor — GPU-aware, Braille charts, per-char label inversion

xlog-prob

Probabilistic inference engines for XLOG

mold-ai-tui

Terminal UI for mold — interactive AI image generation

hardware

A no_std bare-metal hardware abstraction layer — all port I/O, memory and swap allocations are guarded at runtime. Do not consider this dependency stable before x.1.x

soorat

— GPU rendering engine for AGNOS

wbackend

WASMA – Resource-first runtime: CPU-priority, GPU-optional, platform-agnostic

async-cuda

Async CUDA for Rust

apple-mps

Safe Rust bindings for Apple's MetalPerformanceShaders framework on macOS, backed by a Swift bridge

flash-map

GPU-native concurrent hash map with bulk-only API. Robin Hood hashing, SoA layout, CUDA kernels. Designed for blockchain state, HFT, and batch-parallel workloads.

mtl-sys

Low-level Objective-C runtime bindings for Metal

rakka-accel-cuda

GPU acceleration via the actor model. Wraps NVIDIA CUDA libraries (cuBLAS, cuDNN, cuFFT, cuRAND, cuSOLVER, cuSPARSE, cuTENSOR, cuBLASLt, NVRTC, NCCL) as supervised rakka actors with…

mold-ai-inference

Candle-based inference engine for mold — FLUX, SDXL, SD3.5, Z-Image diffusion models

edgefirst-image

High-performance image processing with hardware acceleration for edge AI

xlog-logic

Parser, compiler, and optimizer for XLOG logic programs

find_cuda_helper

Helper crate for searching for CUDA libraries

oxifetch

program that displays key system information, such as OS details, uptime, CPU specs, memory usage, and more. The output includes an ASCII art logo and a quick overview of your machine's current status.

mtl-foundation

Foundation framework bindings (NSObject, NSString, NSArray, etc.)

baracuda-forge

Build-time CUDA kernel compiler for the baracuda ecosystem: nvcc-driven incremental builds, parallel compilation, GPU auto-detection, and CUTLASS / custom git dependency support

neofetch

mold-ai-catalog

Model-discovery catalog for mold — live HF + Civitai proxy with sidecar-driven installed-state

forge-sort

GPU radix sort for Apple Silicon — 4,800+ Mkeys/s, 6 data types, zero-copy

trtx

Safe Rust bindings to NVIDIA TensorRT-RTX (EXPERIMENTAL - NOT FOR PRODUCTION)

beamterm-core

Platform-agnostic OpenGL terminal renderer using glow

vulkan-rust

Ergonomic Vulkan bindings for Rust, generated from vk.xml

sokr

SOKR core — immutable C ABI surface for substrate plugins

glances

A modern Glances-inspired TUI system monitor written in Rust — CPU, memory, GPU, Docker, network, disk, battery, alerts, and container API testing

kronos-compute

A high-performance compute-only Vulkan implementation with cutting-edge GPU optimizations

mtop-tui

Sudo-less system monitor for Apple Silicon Macs: beautiful real-time braille graphs for CPU, GPU, temperature, memory, disk, network, power, and processes

burn-wgpu

WGPU backend for the Burn framework

jolt-platform

Cross-platform battery and power monitoring for jolt

ec-gpu

Traits for field and eliptic curve operations on GPUs

turboplot

A blazingly fast waveform renderer made for visualizing huge traces

runmat-time

Cross-platform time utilities for RunMat (monotonic + wall-clock helpers)

hw_dcmi_wrapper

A safe and ergonomic Rust wrapper for the Huawei DCMI API

zuna-rs

ZUNA EEG Foundation Model — inference in Rust with Burn ML

bl-wgpu-util

Personal utility library for wgpu applications: depth buffers, viewports, image/texture helpers, and a flexible renderer abstraction

mtl-fx

MetalFX bindings for AI upscaling and frame interpolation

blinc_gpu

Blinc GPU renderer - SDF-based rendering via wgpu

cargo-gpu-install

Install rust-gpu and it's required toolchain automatically

concerto-core

Pure routing logic, eviction policies, and memory accounting for Concerto. Zero IO dependencies.

cuneus

A WGPU-based shader development tool

syspeek

A fast and simple system information library for CPU, GPU, RAM, battery, disk, network, packages, and OS info

opencl-sys

OpenCL C FFI bindings for the Rust programming language

zust-vulkan

Vulkan execution helpers for Zust SPIR-V kernels

lattice-fann

Fast neural network primitives for Lattice - tiny models, fast inference

hanzo-pqc

Post-quantum cryptography primitives (ML-KEM, ML-DSA, SLH-DSA) for Hanzo ecosystem

async-tensorrt

Async TensorRT for Rust

tenflowers

Pure Rust implementation of TensorFlow - A comprehensive deep learning framework

beamterm-data

Core data structures and binary serialization for the beamterm WebGL terminal renderer

brutecraber

A blazing-fast, multithreaded hash cracker built with Rust

puma

A lightweight, high-performance inference engine for local AI

meganeura

E-graph optimized neural network training on Blade

vx-vision

GPU-accelerated computer vision for Apple Silicon via Metal compute shaders

neovm-core

Core runtime structures for NeoVM

atomr-accel

Backend-agnostic compute-acceleration core. Defines the AccelBackend trait, AccelRef<T> typed pointers, AccelError enum, and CompletionStrategy — the abstraction layer that lets atomr-accel-cuda (NVIDIA)…

rumus

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

nnl

A high-performance neural network library for Rust with CPU and GPU support

dreamwell-gpu

Dreamwell shared GPU infrastructure — wgpu pipelines, compute, ray tracing, mesh shaders, particles, sprites, procedural generation

mold-ai-db

SQLite-backed metadata store for mold gallery items

ringkernel-wavesim3d

3D acoustic wave simulation with realistic physics, binaural audio, and GPU acceleration

rtop

A system monitor implemented in Rust, Monitors both system activity and GPU activity for NVIDIA GPUs

cuda-rust-wasm

CUDA to Rust transpiler with WebGPU/WASM support

oxicuda-rlhf

RLHF and alignment algorithm primitives for OxiCUDA — DPO, IPO, KTO, ORPO, SimPO, reward modelling, PPO-RLHF

rusthrust

GPU parallel algorithms library inspired by NVIDIA Thrust

ax-driver-display

Common traits and types for graphics device drivers

ha-ndarray

A hardware-accelerated n-dimensional array

burn-cubecl

Generic backend that can be compiled just-in-time to any shader language target

gatenative

execute natively Gate circuits

rustorch-core

Core tensor library for RusTorch

xlog-cuda

CUDA kernel provider, buffers, and interop for XLOG

freecycle

GPU-aware Ollama lifecycle manager for Windows. Monitors for games and GPU-intensive programs, automatically enabling/disabling networked Ollama access when the GPU is available.

ringkernel-cuda-codegen

CUDA code generation from Rust DSL for RingKernel stencil kernels

lumen-server

Embeddable render service primitives and local HTTP server for Lumen

hybrid-predict-trainer-rs

Hybridized predictive training framework with warmup, full-train, predict, and residual correction phases for accelerated deep learning

xlog-runtime

Runtime executor and relation store for XLOG

llmux

Hook-driven LLM model multiplexer with pluggable switch policy

axdriver_display

Common traits and types for graphics device drivers

oxicuda-moe

Mixture of Experts primitives for OxiCUDA — Switch Transformer, Top-K routing, Expert Choice, Soft MoE, load balancing

baracuda-driver

Safe Rust wrappers for the CUDA Driver API (devices, contexts, streams, events, memory, kernels, graphs)

neuronbox-cli

Local ML runner: declarative neuron.yaml, model store, daemon, and terminal dashboard

core-glyph

View tree, Signal<T>, flexbox layout, and flat quad output for the Glyph UI framework

nano-rspow

Hybrid CPU/GPU Nano (XNO) Proof of Work library — Rust core, cross-platform

unmtx-gpu

Micro matrix library for neural networks that uses GPU

runpod

client for the RunPod API

cuda-async

Safe Async CUDA support via Async Rust

par-fractal

Cross-platform GPU-accelerated fractal renderer with 2D and 3D support

oxicuda-continual

Continual and lifelong learning primitives for OxiCUDA: EWC/SI/MAS regularization, PackNet/Piggyback/Progressive architectures, Experience Replay/GEM/A-GEM, DER++, forgetting metrics — pure Rust…

ringkernel-cpu

CPU backend for RingKernel - testing and fallback implementation

rocm_smi_lib

easy to use crate for using rocm-smi from rust

xlog-solve

Solver services used by XLOG exact inference and verification layers

baracuda

Idiomatic Rust wrappers for the NVIDIA CUDA stack (Driver API, Runtime API, NVRTC, cuBLAS, cuDNN, NCCL, NVML, ...). Umbrella crate.

ringkernel-ir

Intermediate Representation for RingKernel GPU code generation

oxicuda-metal

OxiCUDA Metal - GPU compute via Apple Metal API (macOS only)

tensor-crab

Rust-native ML library. No Python. No GIL. Just speed.

flash-rerank-cli

CLI for Flash-Rerank — compile, benchmark, serve, and download models

astrelis-render

Astrelis Core Rendering Module

scenix

Facade crate for the scenix Renderer APIs

luna-rs

LUNA EEG Foundation Model — inference in Rust with Burn ML

oxicuda-recsys

Recommender-system primitives for OxiCUDA — ALS/BPR/NMF, NCF, Two-Tower, DeepFM/AutoInt, SASRec/BERT4Rec, LightGCN/NGCF, MMoE/PLE/ESMM, negative sampling, ranking metrics

oxicuda-quantum

Quantum simulation primitives for OxiCUDA — state-vector simulation, Pauli expectation values, VQE/QAOA, density matrices, Trotter-Suzuki, quantum kernels

wgpu_render_manager

Cached Render/Compute Manager for wgpu (pipelines + bind groups + procedural textures automated)

vkfetch-rs

fetch-program that displays basic information about your vulkan-compatible graphic card(s)!

vyre-primitives

Compositional primitives for vyre — marker types (always on) + Tier 2.5 LEGO substrate (feature-gated per domain)

optirs

Advanced ML optimization and hardware acceleration library (main integration crate)

halldyll_starter_runpod

managing RunPod GPU pods - Provisioning, orchestration & state management

trueno-zram-adaptive

ML-driven compression algorithm selection for trueno-zram

oxicuda-ann

Approximate Nearest Neighbor & vector-search primitives for OxiCUDA — HNSW, IVF, Product Quantization, IVFPQ, LSH, k-NN graph, top-K selection

csep

Cosine Similarity Embeddings Print

archetype_asset

Fast, modular asset system with spatial preloading

zer-compute

Hardware-accelerated backend for zer, pairwise comparison and EM via CUDA, Vulkan, or AVX2

burn-mlx

MLX backend for Burn deep learning framework - native Apple Silicon GPU acceleration

gpu-usage-waybar

display gpu usage in Waybar

keplemon

Expanded functionality for the Standardized Astrodynamics Algorithms Library (SAAL)

steegformer

ST-EEGFormer EEG Foundation Model — inference in Rust with Burn ML

ringkernel

GPU-native persistent actor model framework - Rust port of DotCompute Ring Kernel

cubecl-cuda

CUDA runtime for CubeCL

bevy_gpu_test

A test harness for running GPU compute shaders in Bevy and reading back results for CPU-side assertions

image-colorizer

Never settle for images outside your colorscheme again!

libredr-server

LibreDR is an open-source ray-tracing differentiable renderer

libkrun

A dynamic library providing Virtualization-based process isolation capabilities

ringkernel-metal

Metal backend for RingKernel - Apple GPU support

rsfgsea

High-performance fgsea-compatible preranked Gene Set Enrichment Analysis in Rust

neonwhite_seed_finder

Find and simulate seeds for neon white level rushes

kaio-py

Python bindings for KAIO — Rust-native GPU kernel authoring framework. Exposes the kaio-ops public API (tensor-core matmul, attention, quantized kernels) to Python via PyO3.

feagi-npu-burst-engine

High-performance burst engine for FEAGI neural processing

lumen-engine-ffmpeg

FFmpeg integration for media decode, encode, muxing, and GPU interop in Lumen

libinfer

Rust interface to TensorRT for high-performance GPU inference

vyre-lower

Substrate-neutral lowering: vyre Program → KernelDescriptor consumed by vyre-emit-* crates

mabda

— GPU foundation layer for AGNOS (device, buffers, compute, textures)

crystal-vk

Graphics wrapper for Vulkan

horizon-lattice-render

Graphics rendering backend for Horizon Lattice using wgpu

rakka-accel

Backend-agnostic compute-acceleration core. Defines the AccelBackend trait, AccelRef<T> typed pointers, AccelError enum, and CompletionStrategy — the abstraction layer that lets rakka-accel-cuda (NVIDIA)…

moe-gpu-dsp

MoE-routed GPU signal processing framework — batch cuFFT, kernel dispatch, zero-copy pipelines

volren-gpu

wgpu-based GPU volume renderer for volren-rs