Hosted model catalog
Find public hosted model IDs, lifecycle, context windows, capabilities, and token rates.
The canonical catalog is GET /opchat/v1/models. It is backed by OpenPond's
public model registry and excludes hidden or deprecated entries.
Discover available models
Open the Console's LLM Gateway Models
view, or request the catalog with an API key that has opchat:model:read:
curl https://api.openpond.ai/opchat/v1/models \
-H "Authorization: Bearer $OPENPOND_API_KEY"Use the model ID returned by the catalog in your request. Inspect the available
capabilities, context and output limits, lifecycle, and rates shown for that
model. The openpond-chat alias follows the current default model; use a
specific listed ID when the application needs an explicit model choice.
Catalog availability and prices change. Fetch the current catalog and check the Console before routing production traffic; a copied table of IDs or rates can become stale. A missing or unavailable model is not an invitation to use a hidden provider ID.
The catalog covers hosted models. For trained versions and personalized endpoints, use Models and Serving.