KTX 2.0 (.ktx2) parser and serializer.

ONNXRuntime Node.js binding

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

Path tracing renderer and utilities for three.js built on top of three-mesh-bvh.

A simple mocker for testing threaded development

Whisper.cpp Node.js binding with auto model offloading strategy.

A React implementation of Kaizen UI Foundations. This package is unstyled.

Tailwind CSS plugin for the NVIDIA Design Language, providing utility classes and theme configuration

A JavaScript version of PyTorch with CUDA acceleration

Local RAG System for Claude Code — Hybrid search + Cross-encoder Reranking + 13 MCP Tools + 20 Format Parsers. Zero external servers.

GPU Accelerated JavaScript

High-performance speech-to-text inference addon using NVIDIA Parakeet models for Bare runtime

Sweet Search native binaries for Linux x64 (glibc) with NVIDIA CUDA backend (candle-cuda + flash-attn, SM 7.0+)

Local CUDA LoRA fine-tuning with held-out base-versus-tuned evaluation.

MCP server for remote command execution — run shell/bash commands on remote machines, servers and laptops from your AI agent (Claude, Codex, any stdio MCP client). An SSH / Ansible alternative with no exposed SSH, open ports or VPN; drive machines by AIC-

WebGPU for node

VLA vision-language-action inference addon for QVAC (ggml backend)

API and CLI for compressing images using the ASTC, ETC, ETC2, BCn (DXTn) and BASIS texture compression standard.

WebGPU for Node.js via Google Dawn

Compressing images using the BC1 (DXT1), BC2 (DXT3), BC3 (DXT5), BC4, BC5 & BC7 texture compression standard.