N0X is a local-first AI workbench for people who do not want to upload every file to a SaaS chat app.
Drop in PDFs, notes, docs, code, logs, or CSVs. Ask questions with hybrid search, run local models in the browser with WebGPU, use Ollama for bigger local models, or switch to an OpenAI-compatible cloud provider only when you choose.
Every answer shows the provider path, so it is clear what stayed local and what used cloud. Built for private docs, research, debugging, and reproducible answers.
how does the browser-side inference actually hold up on integrated GPUs, like is it usable on a regular macbook air or are we talking proper discrete card territory?
Love that it actually keeps everything local with browser-side inference, no sketchy data grabs or surprise paywalls. The Ollama and OpenAI-compatible fallback is a smart move for when you need a bigger model without rebuilding the whole workflow.