openjev runs open LLMs locally in the browser via wllama (llama.cpp WebGPU)

ngxson · x · 2026-09-19

ngxson unveils openjev, powered by wllama — the WebGPU/WASM binding for llama.cpp — enabling open-model LLM inference fully in the browser with no backend.

Related event: wllama V3 and openjev bring local LLMs to the browser(3 posts)→

Original post →

More from Infra

Infra channel →