MBio AI Chat Docs
Documentation / AI & RAG

Provider health, model discovery & AI cost tracking

Verify model providers, discover model IDs and measure site-level token usage, latency and estimated spend.

Health checks

Run a minimal request using the website’s published AI configuration. MBio records status and latency without exposing the credential.

Model discovery

Select a stored provider credential to query its OpenAI-compatible /models endpoint. Availability varies by provider and account.

Estimated cost

Enter the provider’s current per-million token price for each model. MBio then calculates estimated spend from API usage metadata. Keep price entries current because provider pricing can change.