How to Use the Prompt Endpoint in Vultr Serverless Inference

The Vultr Serverless Inference prompt endpoint lets you send a single prompt to an AI model and receive a generated response. You can adjust model settings for each prompt and use the responses in your applications.

Follow this guide to use the prompt endpoint on your Vultr account using the Vultr Console.

1. Navigate to Compute and click Serverless Inference.
2. Click the target subscription to open its Manage Serverless Inference page.
3. Click the Prompt tab.
4. Click the Select Model drop-down and select a model.
5. Optionally, set the Max Tokens, Temperature, Top-P, Frequency Penalty, and Seed values.
6. Enter your prompt in the Prompt field and click Prompt.
7. Review the model's response below the prompt settings.
8. Click Reset to clear the prompt and enter a new one.