nodus get models -n nodus lists the inference catalog: each model’s id, operations, context window and the
input and output rates per million tokens a request is quoted at now. -o wide adds aliases and descriptions.
nodus get model gpt-oss-120b -n nodus -o yaml shows the full entry: modalities, capabilities, every rate
(cache reads and writes, audio per second with its minimum, speech per million characters) and the pricing
version receipts record.
A Model is Ready while requests for it are served; the Ready models are exactly the ones
GET /v1/models lists, at the same rates. Select them with --field-selector status.phase=Ready, or one
family with --field-selector spec.family=gpt-oss.