Reprice a model
Request
- Base URL: / (relative, and the document was published on a file host, not at the API’s origin)
- URL: /admin/llm/models/{id}/price
- Auth: HTTP bearer
Path parameters
Request body
Note; required when the model is free (e.g. self-hosted, no cost)
Cache-read rate per 1M tokens (USD), if applicable
Cache-write rate per 1M tokens (USD), if applicable
Input rate per 1M tokens (USD); required when the model has a price
Output rate per 1M tokens (USD); required when the model has a price
PRICED shows the price itself; NO_CHARGE is a deliberate no-cost declaration; UNPRICED shows "No price set"
Response
OK
Model brand declared by an admin; display only
Owning connection's display name
Owning connection id
Context window in tokens
Creation timestamp
Admin-only note: region, agreement, renewal date
Data-handling tier derived from the operator; UNDECLARED until it is set
Human-readable name
Active toggle
Workspace ids shared with; only meaningful when visibility is GRANTED
Model id
Maximum output tokens
Who operates the systems the work is sent to; null until declared
Reasoning effort requested of the model; null sends none, the provider's default applies
Unique slug within the connection
Last update timestamp
Upstream provider model id
Share with all workspaces (PUBLIC) or only selected ones (GRANTED)
Changes
No operation changes recorded.