* [WebLLM v0.2.86](https://github.com/mlc-ai/web-llm) – High-performance in-browser LLM inference engine using WebGPU for hardware-accelerated, serverless model inference.