Unlike Lex (OpenAI primary for everything), Nox only routes to OpenAI
when a reply is actually grounded in retrieved document excerpts —
plain chat and live-data questions stay on the free Cloudflare tiers.
Real streaming, same as the Cloudflare path.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Document current features, how they work, and the fine-tuning plan
A full onboarding reference for this Space, kept in the repo itself
(not a local file) so it's available to any future session regardless
of machine: architecture, every current feature with how it's actually
implemented, the deploy workflow, known limitations, and the ZeroGPU
fine-tuning pipeline plan with its real, unvalidated quota risk called
out explicitly rather than glossed over.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
