AI Studio / Deploy

Deploy gpt-oss-20b

Reasoning model with a 131K context window that runs on a single 16 GB card. Served with Ollama on a GPU you reserve on the Nosana network; you pay from your own wallet and get a private endpoint with a chat playground.

Expect 3 to 10 minutes after a node starts the job before the model answers: the node downloads the image and the weights first, and that time counts toward your reservation.