NVIDIA has open‑sourced its Run:ai platform for managing GPUs on Kubernetes clusters. The code is now under the Apache 2.0 license, which allows anyone to use, modify and redistribute it. The release includes the KAI Scheduler, support for CUDA 13.4 GPU fractioning, quota management features, and a local Kind development environment. This makes GPU scheduling tools more accessible to developers and cloud operators.
Why it matters
Open‑source access lets smaller companies and researchers add fine‑grained GPU scheduling to their Kubernetes clusters without paying for a proprietary solution.
