Azure OpenAI in Foundry Models | Microsoft Azure (original) (raw)

• Azure OpenAI offers both pay-as-you-go pricing and pricing based on provisioned throughput units (PTUs). Pay-as-you-go pricing allows you to pay for the resources you consume, making it flexible for variable workloads. PTUs, on the other hand, offer a predictable pricing model where you reserve and deploy a specific amount of model processing capacity. This model is ideal for workloads with consistent or predictable usage patterns, providing stability and cost control.
• Criteria for evaluating and approving use cases include, but aren’t limited to:
  • Responsible use of the service—using the models as they’re intended to be used.
  • Capacity—ensuring that we’re able to adequately onboard and provide the service to customers.
• Azure OpenAI provides access to many cutting-edge models including GPT-3.5-Turbo, GPT-4 and DALL-E, Whisper, Babbage, and Davinci.
• There are several ways that Microsoft is ensuring responsible use, including:
  • Tools and guidance: Azure OpenAI offers tools to moderate generated content and guidance for safely designing applications.
  • Limited access: The service is currently available with limited access to ensure that responsible AI safeguards are working in practice.
• We guarantee that Azure OpenAI will be available at least 99.9% of the time.