The introduction starts with:

Red Hat Enterprise Linux AI is a platform that allows you to develop enterprise applications on open source Large Language Models (LLMs).

But the only documented way to interact with models is the ilab model chat command, which is hardly a desirable programming interface.

There is an endpoint with an industry-standard API at http://localhost:8000/v1/completions, but it's not clear whether this is a supported interface, or an internal implementation detail of ilab model chat.

If models need to be transferred to RHEL AIIS for serving, this should be mentioned in the documentation (preferably also in the introduction).