QuantClaw is a plug-and-play task-type routing quantization plugin for OpenClaw.

Turn Fake Pixel Art Real — Fast, free, and runs in your browser.

🏆Winning Project | ModelGate is a contract-aware AI control plane that ingests customer contracts, extracts SLA/privacy/routing constraints, and generates an OpenAI-compatible endpoint that automatically routes every request to the optimal model. Simple queries go to cheap models. Complex queries go to premium ones.

An intelligent traffic monitoring system that collects traffic flow and metrics including average speed and vehicle counts for each road. Features real-time data visualization with interactive dashboards and an integrated AI Agent chatbot for querying processed traffic data in real-time.

Open source skill library for AI coding agents to write, optimize, and debug high performance compute kernels across CUDA, Triton, and quantized workloads.

TypeScript implementation of Google's TurboQuant algorithm for near-optimal vector quantization. Zero dependencies, works in Node.js and browsers.

Find which LLMs actually fit on your hardware. Client-side GPU detection, quantization-aware memory estimation, and speed predictions.

A clean-room implementation of Google Research's TurboQuant, with the RaBitQ unbiased-estimator correction. Runs anywhere JavaScript runs; Node, browsers, Bun, Cloudflare Workers, React Native.

Color quantization and dithering in TypeScript.

ReAct coding agent for 4GB laptops. Bonsai 1-bit LLMs, 860MB RAM, zero API fees, works offline. Started because I wanted to build my coffee shop website without paying subscriptions.

Calculate maximum quantized LLM size fitting available RAM

⚡ Compare AI models by Accuracy × Cost × Carbon — RTX 5090 benchmarks reveal 4-bit quantization wastes energy on small models

A web application for the quantization of colors in images.

Wrapture lets you go from a Python-trained model to deployable JavaScript with a single command. It generates TypeScript bindings and a Web/Node-compatible wrapper, using WebGPU/WASM-ready ONNX runtimes.

Fast WebGPU runtime for 1-bit (binary-weight) LLMs in the browser. Bit-exact, zero runtime dependencies.

Running the Multimodal AI Chat App with LM Studio using a locally loaded model

Wrapture lets you go from a Python-trained model to deployable JavaScript with a single command. It generates TypeScript bindings and a Web/Node-compatible wrapper, using WebGPU/WASM-ready ONNX runtimes.

Hardware-aware local-LLM loadout planner: chooses model class, quant, KV-cache, context, and offload policy for your VRAM/RAM, and refuses configs that would silently page to disk.

AI agent toolkit tuned for local LLMs — a pi fork targeting 12B-class models (Gemma @ Q4_K_M) on consumer hardware: enforced token budgets, flat frozen tool schemas, and pre-task ETA + feasibility estimates. See MSLM.md.

Running the Multimodal AI Chat App with Ollama using a locally loaded model