webinarfreeonline
Model providers are switching from per-user to per-token pricing: how an AI gateway helps
Your developers already use coding agents, and your model providers are moving you from per-user licences to per-token contracts. In 30 minutes, live: an unmodified coding agent pointed at one governed endpoint inside the perimeter, a developer routed with a sign-in and a key, the model swapped by configuration, and the spend report that follows a week later — per developer, per model, with a cap. For the platform lead who has to answer for the invoice and the CISO who has to answer for the keys. Then how to start with the gateway on its own, without a programme. Thirty minutes, Q&A included.
Speaking:
Language: English
Agenda
11:00 - Per user to per token: what changed in the contracts, and why it reaches the board first
11:04 - One endpoint, live: a coding agent pointed at the gateway, unchanged (Xavier)
11:10 - Who holds the keys: a developer routed as a person, revoked by removing them
11:15 - A week in: the spend report, per developer, per team and per model, and the cap
11:20 - What a model-only gateway cannot see, and what the same route governs next
11:24 - Starting with the gateway on its own, then Q&A until 11:30