Kueue review
A free, open-source scheduler for Kubernetes batch queues, quotas, and multi-cluster workloads.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
Kueue is a Kubernetes-native job queueing system for batch, HPC, and AI/ML workloads. It is designed for teams running shared compute environments that need admission control, queue priorities, multi-tenant quotas, and resource sharing. Kueue works with Kubernetes Jobs, JobSets, Kubeflow training workloads, Ray workloads, Pods, Deployments, and other Kubernetes integrations. It is open source under the Apache License 2.0 and is installed into an existing, self-managed Kubernetes cluster.
Its strongest fit is workload governance across shared GPU and compute capacity. Kueue supports priority-based queueing with FIFO strategies, fair sharing, cohorts, lending, borrowing limits, preemption, and partial admission. Topology-aware scheduling and GPU and DRA quota accounting add controls for hardware-aware allocation. Prometheus metrics and pending-workload visibility help operators track queued work. MultiKueue extends the model to multi-cluster workload dispatching, while integrations cover Kubernetes Jobs, JobSet, Kubeflow Trainer, KubeRay RayJob, KubeRay RayCluster, AppWrapper, Pods, Pod Groups, Deployments, StatefulSets, and LeaderWorkerSet.
Kueue suits platform teams that want Kubernetes-native batch queueing without replacing Kubernetes core scheduling components. Its free, open-source model is a practical fit for organizations prepared to operate the software within their own Kubernetes clusters. The trade-off is scope: Kueue is centered on batch, HPC, and AI/ML workload admission and quota management rather than general-purpose cluster scheduling. Choose it when multi-tenant quotas, fair sharing, GPU accounting, or multi-cluster dispatch are priorities; consider an alternative when you need a managed service or scheduling centered on non-batch workloads.
Kueue pros and cons
- Where it wins
- Fair sharing, lending, borrowing, and preemption controls
- GPU accounting, topology awareness, and Prometheus visibility
- Supports Kubernetes, Kubeflow, Ray, and multi-cluster workloads
- Where it doesn't
- Requires installation in a user-managed Kubernetes cluster
- Focused on batch, HPC, and AI/ML workload scheduling
- Does not replace Kubernetes core scheduling components
Kueue fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update