AI Endpoints - Billing and lifecycle

Introduction

AI Endpoints is a serverless platform provided by OVHcloud that offers easy access to a selection of world-renowned, pre-trained AI models. The platform is designed to be simple, secure, and intuitive, making it an ideal solution for developers who want to enhance their applications with AI capabilities without extensive AI expertise or concerns about data privacy.

Objective

This documentation provides an overview of the billing and lifecycle management of various AI model categories offered on AI Endpoints. We will cover the cost structure for Large Language Models (LLM), Audio models, Image models, Embedding models, and more. Additionally, we will explain the lifecycle of our models, including model decommissioning and redirecting, ensuring that you have all the necessary information to effectively manage your AI Endpoints.

AI Endpoints model lifecycle

OVHcloud AI Endpoints follows a model lifecycle process to ensure a seamless experience for our customers. This process includes the following steps:

• Track model usage metrics: We continuously monitor the usage of each model on our platform to identify underutilized or obsolete models that may no longer serve the needs of our customers.
• Decommissioning decision: Once a model is identified as a candidate for decommissioning, we make a decision to retire the model. At this point, we communicate the decision to our customers via email and the #ai-news channel of the OVHcloud Discord server. The email is only sent to customers who have been using the model in the last 3 months and informs them that the model will be removed in 3 months to ensure a smooth transition.
• Grace period: After the decommissioning decision is made, a grace period of 3 months is provided for the model we communicated about. This period allows customers to transition to an alternative model, which is provided in our communications. During this time, the model remains accessible via the API but is hidden from our website catalog.
• Removal from active deployment: Once the grace period ends, the model is completely removed from our APIs and returns a 404 error for any new requests.

By following this model lifecycle process, OVHcloud ensures that customers are well informed and prepared for any changes, while also maintaining a lean and up to date selection of AI models.

Billing principles

For detailed information on models, quantization, pricing units, and prices, please visit the AI Endpoints section of the AI and Machine Learning drop-down menu, on the OVHcloud website:

Feedback

Please send us your questions, feedback and suggestions to improve the service:

• On the OVHcloud Discord server

If you need training or technical assistance to implement our solutions, contact your sales representative or click on this link to get a quote and ask our Professional Services experts for a custom analysis of your project.