Show HN: Goku – WebAssembly‑Based LLM Inference and Model Management Tool

Goku is a new project announced on Hacker News under the Show HN label. It provides large‑language‑model inference powered by WebAssembly. The tool leverages the wllama library to run models in the browser. Goku also includes a model manager for handling multiple LLMs. The project is hosted on a personal GitHub page. It aims to make LLMs more accessible without server‑side resources. Users can load and run models directly in the client environment. The author invites the community to test and contribute to the project.