Explore OpenAI GPT-5.5, GPT-5.4, and Codex on Amazon Bedrock

OpenAI GPT-5.5 and Codex Now Available on Amazon Bedrock

OpenAI has announced the general availability of its latest models, GPT-5.5 and GPT-5.4, along with Codex, on Amazon Bedrock as of this week. These advanced AI models are designed to enhance coding capabilities and streamline complex workflows, making them valuable tools for developers and businesses alike. The integration aims to provide high-performance solutions for demanding customer workloads while ensuring data residency compliance.

Enhanced Capabilities of GPT-5.5 and GPT-5.4

The newly released models are tailored for various applications, including coding, reasoning, and complex professional tasks. GPT-5.5 is particularly suited for handling the most challenging customer workloads, while GPT-5.4 offers an optimal balance between performance and cost-effectiveness. Both models can be accessed via the Responses API through Amazon Bedrock’s next-generation inference engine, which prioritizes speed, reliability, and security.

OpenAI’s Codex serves as an AI-powered coding assistant that supports over 4 million developers weekly in writing, refactoring, debugging, testing, and validating code across extensive codebases. With the enhanced capabilities of GPT-5.5 driving inference processes, Codex is expected to introduce a new level of intelligence that is optimized for complex development workflows.

Integration with Development Tools

Developers can utilize Codex through various platforms including the Codex App, Command Line Interface (CLI), and integrations with popular Integrated Development Environments (IDEs) such as Visual Studio Code and JetBrains. All model inference requests are routed through the Responses API on Amazon Bedrock to ensure seamless operation.

For those concerned about data residency requirements, all processing remains within the selected Bedrock Region. This feature allows organizations to maintain compliance with local data regulations while leveraging powerful AI tools without incurring additional costs associated with seat licenses or per-developer commitments.

Getting Started with OpenAI Models on Bedrock

Developers looking to access these models can do so programmatically using the OpenAI SDK or command-line tools like curl. For instance, setting up the OpenAI SDK for Python involves installing it via pip and configuring environment variables for authentication:

pip install -U openai

export OPENAI_BASE_URL="https://bedrock-mantle.us-east-2.api.aws/openai/v1"

export OPENAI_API_KEY=""

export BEDROCK_OPENAI_MODEL_ID="openai.gpt-5.5"

A sample Python code snippet demonstrates how to call the GPT-5.5 model:

import os

from openai import OpenAI



client = OpenAI(

    base_url=os.environ["OPENAI_BASE_URL"],

    api_key=os.environ["OPENAI_API_KEY"],

)



response = client.responses.create(

    model=os.environ["BEDROCK_OPENAI_MODEL_ID"],

    input=[

        {

            "role": "developer",

            "content": "You are a software engineer with excellent AWS cloud knowledge."

        },

        {

            "role": "user",

            "content": "Design a distributed architecture on AWS in Python that should support 100k requests per second across multiple geographic regions."

        },

    ],

    reasoning={"effort": "medium"},

    text={"verbosity": "low"},

)



print(response.output_text)

Alternatively, users can directly call the model endpoint using curl commands for quick interactions.

Key Considerations for Users

• Model Latency: Performance may vary based on several factors including reasoning effort and output length. Users are advised to start with medium effort settings when using GPT-5.5.
• Scaling and Capacity: The new inference engine is designed to manage rapid provisioning across different models efficiently while prioritizing steady state workloads during high demand periods.

Availability and Future Updates

The OpenAI GPT models along with Codex are now available in specific regions: GPT-5.5 in US East (Ohio) and GPT-5.4 in both US East (Ohio) and US West (Oregon). Users can check the full list of regions for future updates regarding availability.

To explore these new capabilities further or get started with them today, interested parties can visit the OpenAI page on Amazon Bedrock or review pricing details available on Amazon’s website.

What This Means

The introduction of OpenAI’s latest models on Amazon Bedrock marks a significant advancement in AI-driven development tools that cater specifically to software engineers’ needs. By providing robust coding capabilities alongside compliance features for data residency, these tools empower developers to tackle complex projects more efficiently than ever before.

For more information, read the original report here.