OpenAI's GPT-5.5 and Codex Reach General Availability on Amazon Bedrock
Summary
OpenAI's GPT-5.5 and Codex coding agent achieved general availability on Amazon Bedrock on June 11, 2026, following a revision to OpenAI's exclusive cloud agreement with Microsoft. This integration allows over 100,000 existing Bedrock organizations to access GPT-5.5, GPT-5.4, and Codex without new vendor relationships or billing paths. API calls through Bedrock inherit AWS-native controls like IAM, VPC, KMS encryption, and CloudTrail for enhanced enterprise governance and security. GPT-5.5 is available in US East (Ohio), while GPT-5.4 is in US East (Ohio), US West (Oregon), and AWS GovCloud (US-West) as of June 3. Codex, used by over 5 million developers weekly, shifts to pay-per-token billing on Bedrock. This move significantly alters the competitive landscape, with AWS now hosting both Anthropic's Claude and OpenAI's models, signaling an end to exclusive cloud partnerships for frontier AI. OpenAI also previewed Daybreak for secure code review.
Key takeaway
For AI Architects evaluating large language model deployments, OpenAI's general availability on Amazon Bedrock simplifies enterprise integration. You can now utilize AWS-native security and governance controls for GPT-5.5, GPT-5.4, and Codex, streamlining compliance and vendor management. This shift allows you to consolidate AI workloads on Bedrock, potentially reducing operational overhead and accelerating secure production deployments without introducing new billing or governance paths. Consider migrating existing OpenAI API usage to Bedrock for enhanced control.
Key insights
OpenAI's GPT models and Codex are now on Amazon Bedrock, integrating enterprise-grade governance and shifting the competitive AI cloud landscape.
Principles
- Cloud-native controls enhance AI model governance.
- Data isolation is key for enterprise AI adoption.
- Exclusive cloud AI partnerships are ending.
Method
OpenAI models on Bedrock route API calls through AWS-native controls for identity, network, encryption, and audit logging, with three inference routing options.
In practice
- Access OpenAI models via Bedrock's Responses API.
- Use Codex with pay-per-token billing.
- Deploy GPT-5.4 in AWS GovCloud (US-West).
Topics
- Amazon Bedrock
- OpenAI Models
- Enterprise AI
- AI Governance
- Code Generation
- Cloud Partnerships
Best for: CTO, VP of Engineering/Data, Investor, Director of AI/ML, AI Architect, MLOps Engineer
Related on AIssential
Editorial summary, takeaway, and curation by AIssential. Original article published by InfoQ.