AWS Welcomes OpenAI & Anthropic in ‘Powerhouse Combination’

Amazon Web Services (AWS) has announced back-to-back deals with Anthropic and OpenAI, integrating their leading AI models into its Bedrock platform to give enterprises wider choice and flexibility in deploying artificial intelligence across their operations.

Anthropic’s Claude Opus 4.1 joins AWS

The first announcement came on 5 August, when AWS confirmed that Anthropic’s Claude Opus 4.1 would be available on Bedrock.

"This is the most powerful Claude model from Anthropic to date and brings enhanced performance for AI agents along with industry-leading coding capabilities," says Matt Garman, CEO of AWS. "For builders creating agentic applications, Opus 4.1 delivers superior performance on long-running tasks with improved planning and detail tracking across complex workflows."

The addition is designed to help organisations develop AI systems capable of handling sustained, complex tasks. This could be particularly valuable for large-scale data centre environments where automation and efficiency are critical.

Kate Jensen, Head of Growth and Revenue at Anthropic, says: "Claude Opus 4 and Claude Sonnet 4 transform AI from a tool into a true collaborator for every person and every team."

OpenAI models debut on AWS

On 6 August, AWS revealed its first-ever integration with OpenAI, bringing the gpt-oss-120b and gpt-oss-20b open-weight models to both Amazon Bedrock and Amazon SageMaker JumpStart.

Matt describes the collaboration as a "powerhouse combination". He adds: "This brings together OpenAI’s leading technology with AWS’s scale, security and deployment capabilities. More importantly, these new OpenAI models on Bedrock are 10x more price performant than the comparable Gemini model, 18x more price performant than DeepSeek R1 and offer 7x better price performance for most workloads than OpenAI o4."

The models give enterprises new levels of flexibility, enabling them to tailor AI to specific needs such as natural language processing or predictive analytics.

"Our open weight models help developers – from solo builders to large enterprise teams – unlock new possibilities across industries and use cases," says Dmitry Pimenov, Product Lead at OpenAI. "Together with AWS, we’re providing powerful, flexible tools that make it easier than ever for customers to build, innovate and scale."

Implications for data centre operators

The integration of these AI models into AWS Bedrock is likely to be of particular interest to data centre operators and customers deploying large-scale AI workloads. 

By offering both Anthropic’s high-performance reasoning models and OpenAI’s customisable open-weight options under a single platform, AWS is enabling more efficient resource utilisation and potentially reducing the need for additional compute infrastructure.

With data centres increasingly tasked with supporting high-demand AI applications, having access to a wider model selection could help enterprises optimise workloads, lower costs and meet sustainability goals by using compute resources more efficiently.

The partnerships also align with AWS’s broader AI investment plans, which include a US$20bn commitment to future AI projects.

Strategic positioning in the AI market

AWS’s dual announcements come as cloud providers compete to establish themselves as preferred platforms for AI development and deployment. 

Offering multiple leading models in one environment allows AWS to appeal to a broad spectrum of customers, from research institutions to hyperscale cloud users.

"Model choice is tremendously important in the AI era," says Matt. "We’re giving customers the freedom to select the right tools for their unique business challenges."

For the enterprise market – and for the data centres underpinning it – AWS’s move provides a single point of access to some of the most advanced AI capabilities currently available, potentially streamlining both development and operational deployment.