Oracle has signed deals to add AI chips from Google, Cerebras Systems and SambaNova to its cloud infrastructure. The Google deal marks its first sale of Tensor Processing Units for use inside another hyperscaler's data centre.
Oracle Cloud Infrastructure is expanding its processor options for running AI models and agents. Cerebras Systems will supply AI inference capacity, while SambaNova's SN50 processor will handle agentic inference.
The deals reflect an effort to reduce dependence on Nvidia and give customers infrastructure suited to different workloads. Oracle is not abandoning Nvidia: it is also adding Vera Rubin NVL72 superclusters and Vera CPUs.
For companies building AI products, the expansion means more hardware choices within Oracle's cloud. The Cerebras service is planned for 2027, and Oracle has not disclosed how much Cerebras hardware it will use.
Google's deal opens a new route for businesses to access its AI processors through another major cloud provider.
More chip suppliers could help customers reduce dependence on Nvidia, but cost savings remain unproven.
The 2027 Cerebras timetable and undisclosed hardware volumes leave its practical availability uncertain for near-term projects.
Written automatically by the Made in 10 Newsroom from the source above.
Google has demonstrated Gemini 4 Argon on heavy tasks spanning 3D rendering, animation and data analysis. The demonstrations rely on server-side processing, moving demanding work away from local hardware.
Google DeepMind has released Genie 4, an AI world model that generates interactive simulations with audio. Scenes can stay consistent for up to 2 minutes, but access is limited to trusted testers before a wider rollout.
Apple is adding an AI assistant to Pages, Keynote and Numbers, built on models from Apple and OpenAI. The apps also support the Model Context Protocol on macOS 27 or later, letting third-party AI agents connect to them.