Model servers

Launch a hosted model server on a GPU from the console, then chat with it directly from the Playground.

A model server runs a model of your choosing on a GPU instance the console manages for you — launch it once, then talk to it right from the console, the same way you would any other Playground model.

Launch a server

1. 1
   Choose a model

   Pick the model to serve — an open-weights model, or one of your own fine-tunes.
2. 2
   Choose a GPU offer

   Pick an offer from the catalog to run the server on, sized for the model you chose.
3. 3
   Launch and watch it load

   The console shows live launch progress and logs while the model loads onto the GPU.
4. 4
   Chat with it

   Once the server reports ready, open it in the Playground and start chatting — no separate integration step needed.

Manage running servers

A Model Servers list shows every server you have running, its status, and its logs. Delete a server when you're done with it to stop billing for the GPU underneath it.