zero-tvm
Models Proof Docs
zero-tvm

An LLM inference engine written by hand in WGSL and TypeScript. Part of a research line on WebGPU kernels by Ahmet Barış Günaydın.

This site

  • Chat
  • Host a room
  • Docs
  • Proof
  • Validate

Research line

  • kernelfusion.dev
  • github.com/abgnydn
  • huggingface.co/abgunaydin
MIT licensed · weights are fetched once from HuggingFace and cached in your browser; prompts and generation stay on your machine in the chat. In a room, a guest’s prompts run on the host’s GPU. github.com/abgnydn/zero-tvm ↗