How to Provision Vultr Serverless Inference

1 min readUpdated 4 November 2026

Set up a Vultr Serverless Inference subscription to host Generative AI models, using the Console, API, or CLI.

Vultr Serverless Inference is an AI model hosting service that scales automatically and does not require you to manage the underlying infrastructure. It supports models for chat and reasoning, image generation, video generation, music generation, speech generation, transcription, embeddings, reranking, and decision applications, with usage-based billing.

Follow this guide to provision a Vultr Serverless Inference subscription using the Vultr Console, API, CLI, or Terraform.

Navigate to Compute and click Serverless Inference.
Click Create Inference to open the Create Serverless Inference panel.
Enter a Label for the subscription.
Review the usage-based billing note. Optionally, click prices for every model to view the Serverless Inference Model Prices sheet, which lists the price of each model by category, such as Language models, Text to speech, Images, Video, and Music.
Select I acknowledge the usage-based cost.
Click Create Inference.
Verify that the new subscription appears in the Serverless Inference list with an Active status.

0 Comments

Start the conversation

Be the first to comment and share your perspective with the community.

Provision a Vultr Serverless Inference Subscription | Vultr Docs