--- title: "What's new in Lens K8S IDE 2026.9.1: Cost Monitor & Optimization — Lens Blog — Lens" description: "Cost Monitoring breaks cluster spend down per namespace with nothing installed on the cluster, plus VictoriaMetrics auto-detection and Azure Government support." url: https://lenshq.io/blog/lens-release-september26-cost/ --- [← Blog](https://lenshq.io/blog/) Lens Release # What's new in Lens K8S IDE 2026.9.1: Cost Monitor & Optimization Yalton Ruiz September 21, 2026 4 min read **TL;DR:** - **Kubernetes clusters use 8% of the CPU they reserve.** You pay for the reservation, not the usage. - **Cost Monitoring** in Lens K8S IDE shows what a cluster costs over 24 hours, 7 days, or 30 days, per namespace, across CPU, memory, storage, and network. - **Nothing to install on the cluster** that consumes more of its resources. - **Optimize cost with AI** hands those figures to your local AI agent, which returns ranked proposals. It changes nothing without your go-ahead. - **Included with Plus, Pro, and Enterprise plans**, and with the 14-day trial. ## Knowing the number usually costs a project Most teams know how much they spend on cloud infrastructure, but far fewer understand which namespaces within a Kubernetes cluster are driving those costs. Getting that level of detail has meant installing something on the cluster to collect it, and that installation eats into the very resources you're trying to account for. Cost is only half of it. Without visibility into cluster efficiency, whether a namespace is over-reserving or about to run out, you don't know what to fix. **Cost Monitoring** in Lens K8S IDE 2026.9.1 shows you what every namespace in your cluster costs, with nothing to install and no added workload. **Ask AI** connects it to the local AI agent you already run, Claude Code, GitHub Copilot CLI, Gemini CLI, Codex CLI, OpenCode, or a custom command, so optimizing your cluster's cost is one click away. ## Cost Monitoring, one click from the cluster **Cost Monitoring** is a new section in the cluster navigator. Open it and the dashboard reports the last 24 hours, 7 days, or 30 days. ![The Cost Monitoring dashboard in Lens K8S IDE: a status bar with total cost, CPU efficiency, memory efficiency and namespace count, a cost over time chart stacking each namespace's hourly cost, and a cost by namespace table with CPU, Memory, Storage, Network, Total and Efficiency columns](https://lenshq.io/blog/lens-release-september26-cost/cost-monitoring-dashboard.webp) The status bar gives you the shape of the problem in one line: - **Total cost** over the window. - **Idle**, the cluster capacity no workload requested, as an amount and a share of the bill. - **CPU efficiency** and **Memory efficiency**, cluster-wide usage against requests. - **Namespaces** in the window, and the **data source** the figures came from. Below it, **Cost over time** stacks each namespace's cost per interval, so a spike has a name attached to it. **Cost by namespace** breaks the spend down across CPU, memory, storage, and network, and sorts on any column. **Efficiency** is the column that matters. At 12.9%, a namespace is reserving eight times what it uses: money back today. Above 100%, it's using more than it reserved: an eviction risk. Lens colours the two differently. Namespace is the smallest unit Cost Monitoring reports: it does not break a namespace down into its workloads. The figures refresh every minute. ## Optimize cost with AI Reading the dashboard tells you where the money goes. It does not tell you what to change. ![The Cost Monitoring dashboard with the Optimize cost with AI button and its tooltip, and the AI tool's reply open in the terminal below, listing ranked savings with their caveats](https://lenshq.io/blog/lens-release-september26-cost/optimize-cost-with-ai.webp) **Optimize cost with AI** sits on the dashboard and hands the figures to your own local AI agent through [Ask AI](https://docs.lenshq.io/k8slens/ask-ai/), the same one you already use and run on your machine. The tool runs against your cluster, so it reads the requests, limits, replicas, and node pools behind the figures itself, workloads consuming the most resources first. It flags what needs to scale because it's short on resources, and what's sitting idle. Then it returns ranked proposals: what to change, what it saves over a month, and what could break. It proposes. It changes nothing until you ask. Together, Cost Monitoring and Ask AI find the resources you've allocated but never use, and turn that into recommendations that bring your cluster's cost down. ## Also in this release **VictoriaMetrics auto-detection.** If your cluster is monitored with VictoriaMetrics, Lens now finds it on its own and shows your metrics. Nothing to set up. **Azure Government and Azure China.** Lens now works with Azure organizations in the US Government and China clouds. Choose your cloud when you add the organization, and Lens signs in there and finds your AKS clusters. Organizations signed in with different Azure accounts now work side by side, so adding a second one no longer stops the first from showing its clusters. ## Get the update Lens K8S IDE 2026.9.1 is available now to download, or upgrade from inside the app. Cost Monitoring is a Premium Feature, included with Plus, Pro, and Enterprise plans and with an active trial. For the full walkthrough, see the [Cost Monitoring documentation](https://docs.lenshq.io/k8slens/cost-monitoring/). ## Get Lens K8S IDE 2026.9.1 [See pricing](https://lenshq.io/pricing/) [Download](https://lenshq.io/download/)