OpenAI models, Codex, and Managed Agents come to AWS (3 minute read)
OpenAI is bringing GPT-5.5, Codex, and managed agents directly to Amazon Bedrock, letting enterprises use OpenAI tools without leaving their AWS environment.
What: AWS and OpenAI announced three capabilities launching in limited preview: OpenAI models (including GPT-5.5) on Amazon Bedrock, Codex configuration with Bedrock as the provider, and Amazon Bedrock Managed Agents powered by OpenAI for deploying multi-step workflow agents.
Why it matters: This addresses a major enterprise adoption barrier by allowing companies to use OpenAI's frontier models while staying within their existing AWS security controls, compliance requirements, and cloud spending commitments, rather than managing separate vendor relationships and data flows.
Takeaway: If you have AWS Bedrock access, you can configure Codex to use Bedrock through the CLI, desktop app, or VS Code extension, and potentially apply usage toward existing AWS commitments.
Deep dive
- OpenAI models including GPT-5.5 are now available on Amazon Bedrock, letting enterprises build AI applications within their existing AWS infrastructure rather than integrating external APIs
- Codex, used by 4 million weekly users for coding, refactoring, test generation, and document work, can now be powered by OpenAI models served directly from Bedrock
- Customers with AWS commits can configure Codex to use Bedrock as the provider through the CLI, desktop app, or Visual Studio Code extension
- All customer data processed through this integration stays within Amazon Bedrock's security perimeter and can count toward AWS cloud commitments
- Amazon Bedrock Managed Agents powered by OpenAI provides infrastructure for deploying production agents that maintain context, execute multi-step workflows, and use tools
- The managed agents service handles deployment, orchestration, and governance complexity, letting teams focus on building useful agents rather than infrastructure
- This partnership solves the "build where you already are" problem for enterprises that need frontier models but can't easily route data outside their AWS environment
- The integration gives developers flexibility across use cases: new AI apps, embedded intelligence in existing products, and complex agentic workflows
- Security-conscious enterprises get a single procurement path from experimentation to production without managing multiple vendor relationships
- The announcement positions AWS Bedrock as a unified control plane for both AWS-native and OpenAI models, competing with multi-cloud AI platforms
Decoder
- Amazon Bedrock: AWS's managed service for accessing foundation models from various providers through a unified API
- GPT-5.5: OpenAI's latest frontier language model (referenced as their "best frontier model")
- Codex: OpenAI's coding assistant product suite used for writing code, refactoring, test generation, and document work
- Managed Agents: AWS service that handles infrastructure, orchestration, and deployment for AI agents that execute multi-step workflows
- AWS commit: Pre-negotiated cloud spending commitment that enterprises make with AWS, often allowing them to apply usage of various services toward the agreement
Original article
OpenAI and AWS expanded their partnership to bring GPT-5.5 and other OpenAI models to Amazon Bedrock, allowing enterprises to build AI applications within their existing AWS infrastructure and security protocols. The collaboration also introduces Codex (OpenAI's coding tool used by 4 million weekly users) on Bedrock and launches Amazon Bedrock Managed Agents powered by OpenAI for deploying multi-step workflow agents in production environments.