Bedrock Adds Explicit Prompt Caching for OpenAI GPT-5.6 Models.

Melanie Li· July 30, 2026 View original

Key takeaways

  • OpenAI GPT-5.6 models (Sol, Terra, Luna) are now available on Amazon Bedrock.
  • Explicit prompt caching offers precise control over prompt segment reuse.
  • This feature helps reduce inference costs and optimize GPT workloads.
  • It enhances efficiency for applications with repetitive prompt structures.

Who benefits

Software DevelopmentE-commerceCustomer ServiceMarketingContent Creation

Summary

Amazon Bedrock now offers OpenAI GPT-5.6 Sol, Terra, and Luna models, featuring explicit prompt caching. This new capability allows precise control over caching and reusing prompt segments, helping to reduce inference costs and optimize existing GPT workloads.

Amazon Bedrock has announced the general availability of OpenAI's GPT-5.6 models, including Sol, Terra, and Luna, expanding its suite of accessible large language models. Alongside this release, a significant new feature, explicit prompt caching, has been introduced. This caching mechanism provides developers with granular control, allowing them to specify which portions of a prompt should be cached and subsequently reused. The primary benefit of this innovation is a substantial reduction in inference costs, as repetitive prompt elements do not need to be reprocessed. It also facilitates the optimization and migration of existing GPT workloads to Bedrock, enhancing efficiency and cost-effectiveness.

Why it matters

Professionals can significantly reduce operational costs and improve the efficiency of their AI applications by strategically using explicit prompt caching, especially for applications with repetitive prompt structures.

How to implement this in your domain

  1. 1Review the Amazon Bedrock documentation on explicit prompt caching for GPT-5.6 models.
  2. 2Identify existing GPT workloads or new applications that can benefit from caching repetitive prompt segments.
  3. 3Experiment with different caching strategies to determine optimal cost savings and performance improvements.
  4. 4Migrate relevant GPT workloads to Amazon Bedrock to leverage these new features.
  5. 5Train development teams on how to effectively utilize explicit prompt caching in their AI solutions.

Original post by Melanie Li

"OpenAI GPT-5.6 Sol, Terra, and Luna are now generally available on Amazon Bedrock, along with explicit prompt caching that gives you precise control over which parts of your prompt are cached and reused. Learn how to get started, set up explicit caching, and migrate existing GPT…"

View on X

Originally posted by Melanie Li on X · view source

Want to go deeper?

Turn these trends into skills with Learnijoy's hands-on AI & tech courses.

Explore courses