Models
The Models page (/models) shows every AI model available to you in your Lumen instance, along with its pricing and current health. Models that are owned by another user and not granted to any of your groups are not listed, and their detail URLs return Not Found without exposing metadata. Disabled, expired, and deleted models are hidden the same way.
Table Columns
| Column | Description |
|---|---|
| Model | Clickable name — links to the model detail page |
| Coins / 1M tokens (input) | Cost per million input tokens |
| Coins / 1M tokens (output) | Cost per million output tokens |
| Total Endpoints | Number of backend servers (underlying servers providing the model) configured for this model |
| Healthy | How many of those servers are currently reachable |
| Last Checked | When the most recent health check ran |
| Status | Current availability (see below) |
Status Badges
| Badge | Meaning |
|---|---|
| ok (green) | All backends are healthy — model is fully available |
| degraded (yellow) | Some backends are down, but at least one is working |
| down (red) | No backends are reachable — model is temporarily unavailable |
| disabled (gray) | Model has been turned off |
| no endpoints (gray) | No backends have been set up for this model |
Models that require a one-time acknowledgment show a required badge, and early-access models show an early access badge — these models may change or be removed at any time, and you must acknowledge that before using them. A model with an end date disappears from this page once the date passes.
Understanding Pricing
Coins are Lumen's internal currency. See the Introduction for a full explanation of how tokens and coins relate. The prices shown here are per million tokens — typical single messages cost a small fraction of that.
Models with higher output rates tend to be more capable or larger. If you are cost-sensitive, compare models on the detail page before running a long batch job.