Cerebras and Oracle Plan a 2027 AI Inference Service on OCI

Cerebras Systems and Oracle plan to offer Cerebras AI inference technology through Oracle Cloud Infrastructure (OCI), giving cloud customers another option for running interactive AI applications. The companies announced the plan Oct. 27 at Oracle AI World in Las Vegas, with availability planned for calendar 2027.

The service is aimed at AI labs, enterprises and developers. OCI’s existing AI infrastructure includes Dedicated AI Clusters in OCI Enterprise AI. The plan does not specify a launch quarter, deployment size, customer commitments or revenue. Product timing and pricing may change.

Cerebras shares rose about 10% during the Oct. 27 trading session.

Inference is the process by which a trained AI model generates outputs from new inputs. Cerebras’ Wafer-Scale Engine combines computing and memory on a single wafer-scale processor, reducing data movement and communication between chips to support faster token generation. “AI inference demands choice,” said Mahesh Thiagarajan, president of Oracle Cloud Infrastructure.

The OCI plan follows Cerebras’ multi-year collaboration with Amazon Web Services, announced in March. In an August update, Cerebras said it expected to bring an inference solution to Amazon Bedrock in the first quarter of 2027. The OCI offering is planned for calendar 2027, with no more specific availability date announced.