KubeAdapt

KubeAdapt monitors Kubernetes costs and helps teams examine spending by cluster, namespace, workload, pod and team. It supports AWS, Google Cloud, Azure, OpenShift and bare-metal Kubernetes, with an agent installed through a Helm chart; the site says setup takes about five minutes. Rightsizing guidance provides per-container CPU and memory recommendations in YAML for GitOps workflows. Recommendations use OOM-aware, bursty-CPU and replica-aware guardrails, but the platform does not apply changes to a cluster. Smart Alerting can send budget, cost-spike and new-workload notices to Slack, email or webhooks. A read-only REST API provides cost, capacity and resource data, using Bearer API keys. KubeAdapt says its agent has read-only Kubernetes permissions and does not collect Secrets, ConfigMaps, environment variables, container logs or network traffic. The free allowance covers the first 100 vCPU across an account. Paid rates range from $0.60 to $1.99 a month per billable vCPU, depending on tier; startup pricing is $0.60 a month per billable vCPU for qualifying companies.

KubeAdapt combines cost views, alerts and rightsizing guidance for Kubernetes environments. Check the GCP discount limitation and network-agent availability if those affect your setup.

KubeAdapt suits teams that need Kubernetes cost visibility and per-container rightsizing guidance across supported environments. It may also fit teams using GitOps or budget alerts to manage costs.

Overview

KubeAdapt is a Kubernetes cost-monitoring and optimization platform for teams managing workloads across cloud and on-premises environments. It is best suited to operators who want detailed cost allocation and guarded rightsizing recommendations while keeping deployment decisions in their own hands. Its breadth across Kubernetes environments and free first 100 vCPU make it worth considering, though GCP discount coverage is incomplete.

Key features

Cost views span clusters, namespaces, workloads, pods and teams, with container-level allocation and cost forecasting. That helps teams trace spend to the level where workloads are configured instead of relying on cluster-wide totals. AWS, Google Cloud, Azure, OpenShift and bare-metal Kubernetes are supported, making KubeAdapt relevant to mixed estates. GCP has a meaningful caveat: on-demand and Spot or Preemptible pricing work without setup, but Committed Use and Sustained Use Discounts are not yet reflected.

Rightsizing produces per-container CPU and memory recommendations as YAML, a natural fit for GitOps review. OOM-aware, bursty-CPU and replica-aware guardrails add operational context, but KubeAdapt does not change the cluster itself. Teams retain control over rollout, at the cost of having to implement recommendations through their existing process. Smart Alerting sends budget, cost-spike and new-workload alerts to Slack, email or webhooks. The read-only REST API exposes cost, capacity and resource data through Bearer API keys.

Installation uses one Helm chart, and the site says setup takes about five minutes. The agent uses read-only Kubernetes RBAC with list and watch permissions only. KubeAdapt says it does not collect Secrets, ConfigMaps, environment variables, container logs or network traffic; aggregated payloads are signed in-cluster and sent over TLS 1.3 with mutual TLS. Those controls are relevant to security-conscious teams, though they do not remove the need to assess the agent against local requirements.

The optional network agent has narrower reach: it requires Linux kernel 5.10 or newer and is currently priced for AWS, with GCP and Azure not supported yet. Teams relying on network intelligence outside AWS should not count on that component.

Pricing

The Free plan costs nothing and covers the first 100 vCPU across the account. It includes all features and email support, requires no credit card, and is the clearest way to evaluate the product or cover a small deployment. Larger deployments move to per-billable-vCPU monthly pricing, with the first 100 vCPU free on standard tiers.

Standard tier 1 costs $1.99 a month per billable vCPU for 0–500 billable vCPU; tier 2 is $1.49 a month per billable vCPU for 500–2,000; tier 3 is $0.99 a month per billable vCPU for 2,000–5,000; and tier 4 is $0.60 a month per billable vCPU for 5,000+ billable vCPU. The declining rate favors larger estates, while the free allowance keeps the first 100 vCPU outside the paid calculation. Startup pricing is $0.60 a month per billable vCPU for companies under three years old, with fewer than 30 employees and under $6M raised; that eligibility makes it a targeted option rather than a general low-volume rate. A free trial is also offered, and email support applies to free accounts as well.

Platforms

KubeAdapt supports API, Linux, self-hosted and web use. That mix suits teams that want Kubernetes-side deployment alongside browser-based cost review or API access.

Who it's for

KubeAdapt is a strong fit for Kubernetes teams that need container-level cost attribution, forecasting and recommendations they can inspect before applying. Its read-only posture and explicit data-handling boundaries may also appeal to teams cautious about agents with write access. It is a weaker fit for organizations that need commitment-discount-aware GCP totals or network intelligence on GCP or Azure.

Pros and cons

• Pro: Cost views reach from cluster to container and team, giving operators useful detail for allocating and investigating spend.
• Pro: Rightsizing guidance is YAML-based and guarded, but changes remain under the team's control.
• Pro: The first 100 vCPU are free across the account, with all features and email support included.
• Con: GCP Committed Use and Sustained Use Discounts are not reflected, which can leave cost views short of actual discounted spend.
• Con: The network agent requires Linux kernel 5.10 or newer and currently supports priced use on AWS only.
• Con: KubeAdapt does not apply rightsizing changes, so teams need a separate review and deployment workflow.

Alternatives

For a broader shortlist, see Kubernetes Cost Management Software. Atmosly Kubernetes Cost Optimization is a better fit for a very small setup that can work within its free plan's single cluster, cloud account, seat and project limits. Krossboard is another freemium option for teams seeking a self-hosted Kubernetes cost tool.

PerfectScale may suit teams whose needs fit its free allowance of up to 200 vCPU per month and unlimited clusters, nodes and pods. CloudBolt offers a free CMP plan capped at 100 resources, which may be preferable when that resource limit fits the deployment. Klustercost is a free open-source option under GPL-3.0, while KubeLedger offers free self-hosted internal use but cannot be sold as a managed service.

Podledger is worth considering for a self-hosted, single-cloud community plan with basic cost dashboards and pod-level visibility. IBM Kubecost is another freemium choice for teams evaluating OpenCost or its free Kubecost tier.

Verdict

Choose KubeAdapt if your team needs granular Kubernetes cost visibility and reviewable, guarded rightsizing guidance without granting an agent write access. Its first 100 vCPU free and broad environment support make it a practical starting point. Look elsewhere if discounted GCP pricing must be reflected accurately or if you need network intelligence beyond AWS.