JSPM

Advanced, lightweight system and OS information library

Classify GPU's based on their benchmark score in order to provide an adaptive experience.

Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level

Native WebGPU implementation for Bun runtime

Prebuilt linux-x64 binaries for bun-webgpu

A multi-threaded, GPU-accelerated, Canvas API for Node

Add-on layers for deck.gl

Prebuilt darwin-arm64 binaries for bun-webgpu

A simple mocker for testing threaded development

Prebuilt darwin-x64 binaries for bun-webgpu

A gl debugger that listens and replays gl (WebGL, WebGL2, and HeadlessGL) gpu commands

Prebuilt win32-x64 binaries for bun-webgpu

High-performance WebGPU charting library

fast matrix math library for WebGPU

A powerful template system for generating WGSL (WebGPU Shading Language) code with support for parameters, conditionals, and multiple output formats including C++ code generation.

Read floating point values back from WebGL

WebGPU renderer scaffold for Geometra

Intelligent CLI tool with AI-powered model selection that analyzes your hardware and recommends optimal LLM models for your system

ESM support for https://github.com/PavelDoGreat/WebGL-Fluid-Simulation.

**GPU accelerated image resizer**

Transpile GLSL to JS

Mesh generation for client-side raster reprojection.

Namespace provider for ComputeSDK - cloud-native sandboxes with optional GPU support

webgpu utilities

Modal provider for ComputeSDK - serverless Python execution with GPU support and zero cold starts

Classify GPU's based on their benchmark score in order to provide an adaptive experience.

Georeferenced image data visualization in deck.gl

GeoTIFF and Cloud-Optimized GeoTIFF visualization in deck.gl

CLI tool for managing vLLM deployments on GPU pods

Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level

Lightning-paid AI inference - monetise any OpenAI-compatible endpoint in 30 seconds

**QVAC SDK** is the canonical entry point to develop AI applications with QVAC.

A C++ native canvas 2D/WebGL component based on gpu opengl glsl shader GCanvas

Port of Python `affine` package for working with affine transformations in JavaScript.

Detect audio and video features in browser

Framework-agnostic WebGPU runtime for fullscreen WGSL shaders with explicit Svelte, React, and Vue adapter entrypoints.

Neuronix GPU Provider Node — earn by contributing compute to the Neuronix network

AI coding agent powered by GPUShare — Turn your terminal into an AI-powered development environment

A fast, GPU-accelerated terminal manager with split panes, project organization, editor sync, and AI agent detection. Native macOS app or browser.

Encrypted GPU inference on untrusted infrastructure

A performance monitor for React Three Fiber. Track FPS, draw calls, memory, and GPU usage.

Detect and rate GPU performance as 'GOOD' or 'BAD'.

Creates PNG thumbnails from 3D STL file with nodejs

**@agpu/bindings** is a set of classes that is made to wrap WebGPU but with modularity and ease of use.

Native bindings for cross-platform GPU texture sharing (Spout/Syphon)

A GPU-accelerated computing library for physics simulations and other mathematical calculations

Shared browser-safe demo runtime and asset helpers for the Plasius gpu-* package family.

Node wrapper around nvidia-smi.

CLI tool for managing vLLM deployments on GPU pods

High-performance WebGPU charting library

GPU-accelerated image filters using WebGL shaders

Native bindings for cross-platform GPU texture sharing (Spout/Syphon)

CLI tool for managing vLLM deployments on GPU pods

A TypeScript ML framework with Rust native backends (CPU, CUDA, WebGPU) — autograd, tensors, and neural network training at GPU speed.

Native bindings for cross-platform GPU texture sharing (Spout/Syphon)

Run Keras models in the browser, with GPU support using WebGL

WebGPU worker runtime with a lock-free job queue for WGSL workloads.

Native bindings for cross-platform GPU texture sharing (Spout/Syphon)

Render Leaflet's tiles as canvas elements.

GPU-Bridge MCP Server — 30 AI services as MCP tools. LLM, image, video, audio, embeddings, reranking, PDF parsing, NSFW detection & more. x402 native for autonomous agents.

OpenClaw gateway watchdog for NVIDIA AI workloads. Alias of openclaw-cli.

Utilities for geographic reprojections.

WebGPU lock-free queue assets for flat and DAG-ready GPU job scheduling.

Native Springs Shaders

Physics integration bridge package for Plasius GPU renderer migration.

Advanced lighting WGSL modules and planning profiles for @plasius/gpu-worker.

A node.js embedding tool with optional GPU acceleration

Experimental layers for deck.gl

Opt-in GPU debug instrumentation for tracked memory, dispatch, queue, and frame-budget metrics in Plasius WebGPU runtimes.

CLI tool for managing vLLM deployments on GPU pods

CLI tool for creating Modal sandboxes with GitHub repositories

ObelyZK SDK — verifiable ML inference on Starknet with recursive STARK proofs

Zero-config CLI for monitoring your local LLM hardware, runtimes, and model compatibility

Lambda Labs provider for ComputeSDK - GPU-accelerated code execution for ML and AI workloads

Native CLI worker for the c0mpute.ai distributed inference network. Runs LLM inference via ollama and connects to the orchestrator via Socket.io.

High-performance SGP4 GPU/WebGPU/WebGL WASM wrapper.

n8n community node for deAPI - AI image generation, video generation, transcription and prompt optimization

Modal sandbox provider for [Evolve SDK](https://www.npmjs.com/package/@evolvingmachines/sdk).

Local-first LLM inference engine with OpenAI-compatible API, MCP tools, and hardware-aware optimization

PyTorch like deep learning inferrence library

GPU fragment shaders on live, interactive DOM elements

CLI tool for managing vLLM deployments on GPU pods

Evolve SDK lets you run and orchestrate terminal-based AI agents in secure sandboxes with built-in observability.

WebGPU for Node.js via Google Dawn

A fast, productive library for 2D cross-platform development.

Production-ready CLI tool for hardware detection and publishing

Framework-agnostic WebGPU renderer runtime for Plasius projects.

CLI tool for managing vLLM deployments on GPU pods

CLI tool for managing vLLM deployments on GPU pods

Windows AI 开发环境一键诊断与修复工具 - 扫描 37 个维度，反馈收集，社区悬赏，经验值系统

Advanced speech-to-text transcription tool using OpenAI Whisper with GPU acceleration support

Particle effect WGSL modules and helpers for @plasius/gpu-worker.

Unity integration utilities for OneJS - esbuild/PostCSS plugins for USS, GPU compute, and more

Local GPU translation MCP server — TranslateGemma via Ollama, 57 languages, zero cloud dependency

Agnostic GPU-accelerated image editing engine powered by Pixi.js

CLI tool for managing vLLM deployments on GPU pods

Framework-agnostic multi-camera control runtime with parallel multiview planning.

The full EPSG projection database, compressed to 309kb for the web.

ObelyZK CLI — prove ML models and verify on Starknet from your terminal

Real-Time GPU Texture Codecs for the Web

Node.js client for Vast.ai API - Rent GPUs for machine learning and AI workloads

Daeva — local GPU pod orchestrator for AI workloads

Detect and report browser and hardware features

TypeScript SDK client for the QNSP AI orchestration service. Provides secure AI workload management, enclave inference, and encrypted training.

Zarr visualization in deck.gl.

A Rust inference engine that runs a swarm of tiny specialist models that collaborate and learn continuously — on your GPU.

OpenGL ES 3.0 bindings for Node.js via EGL — native on Linux/ARM, ANGLE on macOS/Windows

Command-line tool for deploying and managing AI workloads across edge and cloud compute providers.

Cross-platform GPU/CPU tuning + monitoring for local AI runtimes

llama.cpp bindings for Bare

M-POLY-VTD AI Engine (Loom v0.75.0) — 21 Numerical Types, WebGPU, DNA Evolution, NEAT

Provider-abstracted SDK for routing AI inference across edge and cloud compute — io.net, Akash, Acurast, Fluence, Koii, AWS, GCP, Azure, Cloudflare, and Fly.io.

Chroma key a video/image/canvas element in real time using the GPU

Real-time macOS Activity Monitor in your terminal. Live CPU, memory, GPU, disk, network, battery, and process stats with spatial navigation. Built with terminaltui.

A multi-threaded, GPU-accelerated, Canvas API for Node

TypeScript/Rust/Bun wrapper for Modal GPU cloud - covers all patterns from video2robot/mjlab hackathon

Parallel video rendering with live dashboard, GPU auto-detection, checkpoint system, and stream-copy concat. MCP server + Claude Code skill + CLI. The most powerful free ffmpeg rendering toolkit.

High-performance WebGL2 terminal renderer with sub-millisecond render times

Kill Switch CLI — monitor cloud spending, kill runaway services from the terminal

High-performance CUDA to WebAssembly/WebGPU transpiler with Rust safety - Run GPU kernels in browsers and Node.js

A GPU-accelerated Canvas Graphics API for Node

Run LLMs locally in Node.js with an OpenAI-compatible API

Complete Agentic GPU Infrastructure for Claude Code — 192 MCP tools: Full training lifecycle, inference deployment with cost guardrails, Ray cluster management (Wide-EP, disaggregated P/D, parallelism strategy), vLLM lifecycle (start/stop/sleep/wake/infer

Write GPU kernels in TypeScript, run on any GPU

OpenAI-compatible inference endpoint that automatically routes requests to the fastest, cheapest available compute backend.

Modern GPU Compute Framework in Javascript

Infovis layers (non-geospatial) for deck.gl

WebGPU for Node.js via wgpu-rs (modern, lightweight alternative to Dawn)

React Native / Expo SDK for calling AI inference endpoints from iOS and Android apps — with automatic routing, failover, and secure key storage.

Detecta seu hardware e mostra quais modelos de IA voce consegue rodar localmente

JavaScript SDK for GhostStream transcoding service

WebGL2 implementation for Node.js on top of native-gles

Using the Vast.AI API in an n8n environment.

Create GPU-rendered Svelte components

Node.js bindings for whisper.cpp - fast speech-to-text with GPU acceleration

A high-performance image carousel library powered by WebGL with stunning visual effects

High-performance Node.js native bindings for Facebook FAISS with float and binary indexes, CLI tooling, optional GPU migration hooks, async operations, persistence, and vector utility helpers.

Lightweight WebGPU vector rendering engine

WebGPU rendering engine for high-performance visualization

GPU-accelerated multibeam sonar rendering, phase correlation, and mosaic stitching via WebGPU

High-performance GPU-accelerated text layout and rendering library

Read HWiNFO64 sensor data from Windows shared memory — fully typed TypeScript library

CLI tool for managing vLLM deployments on GPU pods

Prettier plugin for WGSL (WebGPU Shading Language)

ObelyZK CLI — prove ML models and verify on Starknet from your terminal

High-performance charting library using WebGPU and D3.js

Tensor computation and autograd with WebGPU acceleration

CLI tool for managing vLLM deployments on GPU pods

CLI tool for managing vLLM deployments on GPU pods

WGSL minifier for WebGPU shaders - WebAssembly build

NumPy for the browser GPU — zero shaders, zero dependencies

High-performance GPU-accelerated image filters for React Native with New Architecture support

GPU accelerated BLAS for node and the browser

Thousands of clustered lighting for Three.js

WebGL2-Powered library for Graph Visualization

Advanced, lightweight system and OS information library (with updated dependencies)

Device performance monitoring, profiling, benchmarking, and adaptive quality library for React Native

A complete deep learning library for JavaScript — Dense, CNN, RNN/LSTM/GRU, Attention, GPU acceleration, and ESP32 deployment. By CavyIoT.

Add-0n geospatial layers for deck.gl

Intelligent CLI tool with AI-powered model selection that analyzes your hardware and recommends optimal LLM models for your system

node.js but with hardware-accelerated WebGL hardware GPU a

GPU-accelerated halftone effect with WebGL — lightweight, Webflow-friendly

AAA-quality GPU particle system for Three.js with Niagara/Unity VFX Graph-like features

Decentralized compute and data marketplace for AI agents with spot pricing

CLI for nvGraph, which is a GPU-based graph analytics library written by NVIDIA, using CUDA.

Compare your AI inference costs across providers. Audit what you spend on LLM, embeddings, image gen, TTS, transcription, OCR, video, reranking — then benchmark a unified API with 30 services and $1 free trial.

Public API Tools. Completely Unlimited Free APIs.

A binding to OpenCL for Node.JS

Official TypeScript/JavaScript SDK for GPU renters on DC1 — Saudi Arabia's compute marketplace

Modal GPU Cloud MCP server - manage GPU workloads, training, and robot inference

Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level

React Component using Particle System to animate data-flow

BitSage STWO Prover SDK - Unified interface for GPU, TEE, and WASM proof generation

Thunder Compute MCP server — manage GPU instances from AI coding agents

CLI tool for managing vLLM deployments on GPU pods

GPU-accelerated 2D rendering SDK for web. Unit coordinate system, multi-viewport, resolution scaling.

A declarative particle system for React Three Fiber. It uses an FBO Simulation to provide incremental updates to particles, allowing for an intuitive API with use of forces. **Note**: This package is experimental and may introduce breaking changes in futu

WebGL shim for node

React Desktop UI Framework — native GPU-rendered apps with JSX

WebGL2-Powered library for Graph Visualization

Zig-first WebGPU runtime for Node.js, Bun, and Deno

Hardware-aware LLM model scoring MCP server — detects GPU/RAM/CPU and ranks 900+ AI models by hardware fit

Search, hash, sort, fingerprint, and fuzzy-match strings faster via SWAR, SIMD, and GPGPU

CLI for managing RunPod GPU cloud infrastructure. Self-describing for AI agents via --ai flag.

WebGPU × WebAssembly rendering and computing engine for scientific workloads in the browser.

GAME — Visual Experience Language. Compiles .game DSL to zero-dependency WebGPU/WebGL2 Web Components.

Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level

CLI tool for managing vLLM deployments on GPU pods

Production-ready CLI tool for hardware detection and publishing

MCP server for Forge GPU kernel optimization — generate and optimize Triton/CUDA kernels on real H100/A100 GPUs from any AI coding agent

OwnRig MCP Server — AI hardware compatibility data for Claude, ChatGPT, Cursor, and any MCP-compatible assistant. 50 models, 25 devices, 9 machines, 663 compatibility entries.

A fast, network-connected, differentiable tensor library for TypeScript (and JavaScript). Built with bun + flashlight for software engineers and researchers alike.

WebGPU native bindings for win32-x64-msvc

System and OS information library for Node.js, and SystemInfo API.

WebGL2-Powered Visualization Components for Graph Visualization

Search and compare GPU and LLM pricing from your terminal

Performance benchmarking tool for headless browser automation. Measure CPU, GPU, timing, and screenshot performance across configurations.

Lambda Cloud CLI tool with interactive TUI and MCP server for GPU instance management

Classify GPU's based on their benchmark score in order to provide an adaptive experience.

High-performance CUDA to WebAssembly/WebGPU transpiler with Rust safety - Run GPU kernels in browsers and Node.js

WebGPU native bindings for linux-x64-gnu

Low-level Node.js bindings for llama.cpp. Core library for running LLM models locally with native performance and hardware acceleration support.

GPU accelerated neural networks made simple for Javascript

AI-powered CUDA kernel optimization CLI

Create thumbnail from 3D STL file. Creates beautifully rendered png and jpeg output server-side with no GPU from ASCII and Binary STL's.

UMAP with HNSW kNN and WebGPU-accelerated SGD

CLI tool for managing vLLM deployments on GPU pods

Thunder command line utility

GPU based particle system for three.js. Heavily based on tdl library (https://github.com/greggman/tdl)

CLI tool for managing vLLM deployments on GPU pods

Auxot GPU worker CLI — connects your GPU to an auxot-router

MCP server for Forge GPU kernel optimization — generate and optimize Triton/CUDA kernels on real H100/A100 GPUs from any AI coding agent

WebAssembly-based shader language converter and validator. Convert between GLSL, WGSL, SPIR-V, Metal, and HLSL in the browser.

Advanced, lightweight system and OS information library

BitSage Network SDK - Client library for distributed compute, privacy swaps, and Obelysk encryption

SSH proxy dashboard - monitor, file manage and terminal for remote servers

Fused WebGPU compute kernels for transformer inference — 4,081x avg on Apple Silicon, 826x on phones (487 devices tested)

High-performance WebGPU particle engine with compute shaders, GPU instancing, GLB model support, and skeletal animations

execute shader programs

Nebius AI Cloud API client for TypeScript - compute instances, disks, networks, GPU clusters

GPU particle engine — WebGL shaders + React Three Fiber

A JavaScript neural network library with hardware acceleration support

System monitor dashboard widget for Node-RED Dashboard 2.

Node-RED GPU functions

OpenAI-compatible inference endpoint that routes requests through Phonix's decentralised GPU and TEE compute network.

CLI tool for managing vLLM deployments on GPU pods from Prime Intellect, Vast.ai, etc.

Lightweight WebGL wrapper for simple, beautiful shader effects

CLI tool for managing vLLM deployments on GPU pods

Auxot CLI — generate keys and deploy to Fly.io in seconds

Process fragment shader on a rectangular canvas, webgl-less

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

WebGL2-Powered React Components for Graph Visualization

Free cloud GPUs for learning CUDA

CLI tool for managing vLLM deployments on GPU pods

A modern, type-safe FFmpeg wrapper for Node.js with zero dependencies

NodeJS GPU Accelerator

GPUComputationRender module for Three.js with ES6 compatibility

tiny gpu powered charts

React Component for a Particle System Canvas

Fast f16 to f32 decompression using WebGPU

WebGL2-Powered Visualization Components for Graph Visualization - Modernized for 2025

A React hook that gathers graphics card information

Local LLM inference for Node.js. GPU-accelerated. Zero config. Works standalone or with Vercel AI SDK.

GPU-accelerated Deep Recurrent Neural Networks and LSTMs in Typescript. Ported, object-oriented and refactored version of Andrej Karpathy's recurrent-js (https://github.com/karpathy/recurrentjs)

NVIDIA SMI wrapper to get NVIDIA GPU info

## About <a name = "about"></a>

Layers for compact timeline visualizations in deck.gl

A WebGL library simulating fluid dynamics and motion of solid objects within the fluid.

Ultra-fast WebGPU graph visualization library