Introducing explicit prompt caching for OpenAI GPT-5.6 models on Amazon Bedrock
OpenAI GPT-5.6 models launch on Amazon Bedrock with explicit prompt caching at 90% discount.
“Cached input is billed at a 90 percent discount (see the Amazon Bedrock pricing page) and stays available for reuse for 30 minutes.”
OpenAI's GPT-5.6 model family (Sol, Terra, Luna) is now generally available on Amazon Bedrock, deepening the OpenAI-AWS partnership by bringing frontier models into AWS's security and billing ecosystem. A standout feature is explicit prompt caching, which lets developers mark specific prompt sections for reuse at a 90% token cost reduction — particularly valuable for agentic workflows where system prompts and tool definitions repeat across many calls. This signals growing cross-cloud AI model distribution and a competitive push on inference cost for production agentic systems.