| Name | Owner | Status | AI usage (7 days) |
|---|
Settings here apply to every shop and take effect on the next AI message — no redeploy needed.
Which Claude model each AI job uses. Cheaper models cost less per message; more capable ones handle tricky conversations better. Use Test to check a model works with your API key before customers use it (sends one tiny request, a fraction of a cent).
Costs are estimates from Anthropic list prices (per 1M tokens) and a typical chat message (~3.7k input tokens) — check current pricing.