Explore how OCI supports GPU computing for ML model training
OCI Compute enables machine learning (ML) engineers to build next-generation AI models for any workload.
![High throughput bare metal GPU clusters](https://www.oracle.com/oce/dc/assets/CONTB007FB0844134FD289E447AF0E6EAFA6/native/lyft-logo-pine.png)
High throughput bare metal GPU clusters
OCI’s bare metal NVIDIA GPU instances offer a high performance platform for applications that rely on machine learning, image processing, and high performance computing jobs.
![Cluster support](https://www.oracle.com/oce/dc/assets/CONT2E2D72D6252040A39F3EEFD8CBD41544/native/ndcm-analytics.png)
Cluster support
Get industry leading support throughout the entire lifecycle of your ML model training. OCI sales and engineering will ensure your clusters are running efficiently at scale.
![High-speed RDMA cluster networks](https://www.oracle.com/oce/dc/assets/CONTCB547B7431434B05AB6E2E3BF5A035C3/native/sdrs-visually.png)
High-speed RDMA cluster networks
With NVIDIA RDMA NICs achieving up to 1600 Gbps internode bandwidth, OCI provides performant networking to significantly reduce your training times.