ngxson/wllama
WebAssembly binding for llama.cpp that enables running LLM inference directly in web browsers using WebGPU and WASM SIMD.
Not currently ranked — collecting fresh signals.
star history
wllama provides WebAssembly bindings for the llama.cpp C++ library, allowing large language models to run entirely client-side in web browsers without any backend server. It leverages WebGPU for GPU acceleration and WebAssembly SIMD for multi-threaded CPU inference. The project supports multimodal inputs including images and audio, tool calling capabilities, and exposes an OpenAI-compatible API.
Frequently asked
- What is ngxson/wllama?
- WebAssembly binding for llama.cpp that enables running LLM inference directly in web browsers using WebGPU and WASM SIMD.
- Is wllama open source?
- Yes — ngxson/wllama is open source, released under the MIT license.
- What language is wllama written in?
- ngxson/wllama is primarily written in TypeScript.
- How popular is wllama?
- ngxson/wllama has 1.2k stars on GitHub.
- Where can I find wllama?
- ngxson/wllama is on GitHub at https://github.com/ngxson/wllama.