OpenAI GPT-5.6 Models Now Available on Amazon Bedrock with Explicit Prompt Caching

OpenAI and AWS have announced the general availability of GPT-5.6 Sol, Terra, and Luna models on Amazon Bedrock, introducing explicit prompt caching for enhanced efficiency.

New Model Capabilities

The GPT-5.6 family offers three tiers: GPT-5.6 Sol for complex reasoning, GPT-5.6 Terra for balanced workloads, and GPT-5.6 Luna for high-volume tasks. These models come with pay-per-token pricing, AWS security, and governance controls.

Explicit Prompt Caching

Explicit prompt caching allows users to control which parts of their prompts are cached and reused, billed at a 90 percent discount and available for 30 minutes. This feature is particularly beneficial in agentic workflows.

Migration and Setup

Developers can migrate existing GPT workloads to the new models and set up explicit caching to optimize performance and reduce costs. Detailed instructions are available in the Amazon Bedrock documentation.

Market Context

Microsoft, a major cloud provider and stakeholder in both OpenAI and Anthropic, is competing more openly with these AI labs as they expand into applications and agentic infrastructure.

"Using multiple models and avoiding reliance on frontier AI labs for the app layer is safer," said Microsoft CEO Satya Nadella.

"Microsoft reported $90 billion in revenue and net income of $35.8 billion for the quarter, and $331.8 billion in revenue with a net income of $133.7 billion for the fiscal year."