What is ModelRun
ModelRun offers managed GenAI endpoints for open-source or customer-supplied models. The service positions itself around instant deployment, dedicated endpoints, usage dashboards and managed GPU infrastructure.
Deployment model
Customers can choose from a curated model library or upload a model, then receive an API endpoint and key. The official site also describes batch-optimised inference, no cold starts on dedicated endpoints and transparent input/output metering.
Pricing
As of August 16, 2026, ModelRun's official public page does not expose a current named model catalogue or numeric per-model rates. It advertises free access through OpenRouter and a sign-up route for managed deployment, but buyers should confirm the live model and rate card before use.
Official source: ModelRun product overview.