Amazon Bedrock is the platform for building and running AI applications and agents at production scale. OpenAI models on Bedrock run on Amazon Bedrock’s next-generation inference engine, built for high performance, reliability, and security.

GPT-5.5 grasps your intent faster and handles multi-step tasks autonomously, excelling at writing and debugging code across large code bases, analyzing data, generating documents and spreadsheets, and operating software across multiple tools until a task is complete. The improvements are most significant in agentic coding and knowledge work, where real progress depends on sustaining context and taking action over time.

Both GPT-5.5 and GP-5.4 are built for complex, multi-step tasks and are available in the Amazon Bedrock model catalog today. You can call them through the Responses API on Amazon Bedrock and pay the same per-token rate as direct from OpenAI with no additional fees.

Bedrock’s inference engine gives you your own isolated queue with automated capacity management, so your performance stays predictable, even under heavy load. As each request runs, its full state is captured durably and continuously, so if hardware fails or a node restarts mid-call, your request picks back up where it left off instead of starting over. Every call inherits the governance controls you already use across AWS: IAM permissions, VPC and PrivateLink isolation, KMS encryption, and AWS CloudTrail audit logging. Your prompts and responses are not used to train models and are not shared with model providers. These protections extend to GPT-5.5 and GPT-5.4 on Amazon Bedrock.