Local cross-encoder re-ranker for Virage (ONNX, no API key required)
Background Removal plugin for the CE.SDK editor
StarCraft 2 Terran Adjutant voice synthesis powered by onnxruntime-node + espeak-ng.
Consolidated ONNX plugin providing object detection, face detection and recognition, license plate detection with OCR, and CLIP semantic embeddings.
Core inference engine for TimesFM — zero-shot time series forecasting
Run Ultralytics YOLO models in the browser on WebGPU (WebAssembly, ONNX Runtime Web via ort-web).
High-performance ONNX Runtime bindings for React Native with hardware acceleration support (NNAPI, CoreML, XNNPACK)
Semantically create chunks from large texts. Useful for workflows involving large language models (LLMs).
A Javascript library for running ONNX models on browsers
Local-first image annotation tool for agriculture with AI-powered segmentation (SAM). Annotate crop regions, plant parts, and disease areas. Exports to YOLO, COCO, and LabelMe formats. Runs entirely on your machine — no cloud, no uploads, no account neede
Mimir add-on for plug-and-play Edge-quality and offline text-to-speech audio summaries.
A high-performance React Native library for text-to-speech on iOS and Android
Self-hostable, OpenAI-compatible Whisper speech-to-text API server you can run anywhere with npx. Local inference via whisper.cpp or ONNX (transformers.js).
A collection of AI utilities, inference pipelines, and development tools for building intelligent Unity applications using Unity Sentis, ONNX models, and custom machine learning workflows.
Umbrella entry for wetron - parses every supported model format via parseModel and re-exports the parser-specific APIs
ONNX model parser - extracts graph structure (nodes, edges, shapes) from .onnx files
Local Transformers.js embedder for Kura (bge-m3). Runs ML models in JS with no Python.
On-device face liveness, gesture verification and age estimation for React Native using ONNX Runtime
Node-first OCR interface with support for Tesseract.js, ONNX (PaddleOCR), and HappyVertical AI-backed vision OCR
A lightweight, real-time attention tracking and engagement analytics library for detecting focus, distraction, drowsiness, gaze behavior, and psycho-emotional states from webcam video.
Complete local model SDK for EASBot - Transformers.js + ONNX implementation supporting both Language Model (text generation) and Embedding (text embedding) with full AI SDK v2 compatibility
Qubit.ai — Generative AI with Pyodide + NeuroQuantum (quantum-inspired text generation without external APIs)
FastEmbed ONNX-based local embedding provider for @vivantel/rag-core
AI coding agent powered by open-source models (Ollama/vLLM) — interactive TUI with agentic tool-calling loop
On-device audio analysis for React Native — WAV decoding, DSP (RMS, transients), ONNX ML event detection, prominence scoring, and highlight selection via Nitro Modules