Cerebras will bring its fast AI inference service to Oracle Cloud.

Cerebras and Oracle have agreed to bring Cerebras AI inference to Oracle Cloud Infrastructure, targeting AI labs, enterprises and developers building interactive AI applications.

The addition will expand Oracle's cloud portfolio and is planned to be available to Oracle Cloud Infrastructure customers in calendar year 2027.

Inference speed has become a key issue for AI developers as applications move beyond single prompts to multi-step tasks. Each additional model call can add delay, affecting the responsiveness of AI agents that need to reason, check outputs and take action across several stages.

The new offering will sit alongside Oracle's existing AI infrastructure, including Dedicated AI Clusters in OCI Enterprise AI. For Cerebras, the deal extends the reach of its inference technology through a large public cloud provider.

The service is aimed at workloads where response time matters, particularly agent-based applications that carry out longer reasoning sequences or more complex chains of activity.

Cerebras' technology centres on its Wafer-Scale Engine, a processor design that combines compute and memory on a single wafer. According to the company, that approach reduces data movement and communication between chips, supporting faster token generation during inference.

The deal reflects a broader shift in the AI market from training large models to running them efficiently in production. As businesses try to deploy AI tools for customer service, software development and internal automation, the speed and consistency of responses have become practical as well as technical concerns.

Andrew Feldman, Co-Founder and Chief Executive Officer of Cerebras, framed the partnership around demand for quicker responses from AI systems.

"People want AI that can do valuable work quickly," said Andrew Feldman, Co-Founder and Chief Executive Officer of Cerebras. "Every step an agent makes requires inference, whether it's reasoning, checking its work, or taking action. The faster the inference, the more work AI can do without making users wait. Bringing Cerebras fast inference to Oracle Cloud Infrastructure gives AI labs, enterprises, and developers the speed to build agents that do more, faster."

Oracle has been expanding the range of compute options it offers to cloud customers as competition intensifies in AI infrastructure. Cloud providers are increasingly combining their own systems with specialist hardware from outside suppliers to appeal to customers with different performance and cost requirements.

That has created room for chip and system companies focused on specific parts of the AI stack, including inference. While much of the recent spending boom has centred on training hardware, inference is emerging as a distinct battleground because it directly affects the running cost and user experience of deployed AI services.

Mahesh Thiagarajan, President of Oracle Cloud Infrastructure, said customers wanted a wider set of infrastructure choices for AI workloads.

"AI inference demands choice," said Mahesh Thiagarajan, President of Oracle Cloud Infrastructure. "As agents take on more complex work, customers demand compute that matches each application's performance needs. Bringing Cerebras' inference to Oracle Cloud Infrastructure will add another ultrafast and deterministic option to our broad compute portfolio, helping customers build agents that respond quickly and tackle demanding tasks."

The move also underlines how cloud groups and AI hardware suppliers are working more closely as enterprises test which combinations of model, infrastructure and software are best suited to production use. For Oracle, adding Cerebras gives customers another route for inference-heavy applications. For Cerebras, it offers access to Oracle's cloud customer base at a time when buyers are weighing alternatives for running AI systems at scale.

Cerebras Inference on OCI is planned to be available in calendar year 2027.