CostAnalyst
Blog / Buyer guides 6 min read

Densify Pricing Now Kubex at $1 per vCPU a Month With a 60-Day Trial

By the CostAnalyst team

Spend Console
Cloud + SaaS
Loaded AWS GCP Azure SaaS
Find savings in
Identified

projected this month if unattended

Spend by team

Budget forecast

Projected EoQ $124k
With savings
No account access ยท money never moves

Densify is now Kubex, and Kubex prints a price: $1 per vCPU per month for its Kubernetes Resource Optimization plan, with a 60-day free trial. A cluster of 500 vCPUs costs about $500 a month, or $6,000 a year, and 1,000 vCPUs costs about $1,000 a month. GPU and SSO features sit in a custom Enterprise plan priced per vCPU for Kubernetes and per GPU for accelerators.

Most comparison posts still say "Densify pricing is custom". That was true under the old name. We read the current Kubex pricing page and its machine-readable pricing file on 16 November 2026, and compared them with the AWS Marketplace tiers we logged earlier. The vendor states that the rename was a name change and not a change of ownership, so a Densify quote from before 2026 and a Kubex quote today come from the same company.

How much does Densify (Kubex) cost?

Two plans are listed. Kubernetes Resource Optimization is usage-based, billed per vCPU per month, and covers pod scaling, node optimization, a bin packer, HPA tuning, a sizer for new containers, and cloud instance and RDS optimization. Enterprise K8s and GPU Optimization adds GPU and AI infrastructure features, single sign-on and third-party observability ingestion, and is quote only.

PlanPriceUnitAdds
Kubernetes Resource Optimization$1per vCPU per monthFull Kubernetes suite, cloud instance and RDS optimization, UI, CLI and MCP access
Enterprise K8s and GPU OptimizationCustomper vCPU for Kubernetes, per GPU for GPUGPU Resource Optimizer, MIG Planner, SSO, third-party metrics

The page does not say whether the vCPU count is provisioned, requested or averaged over the month. That single definition can move a bill by a third on a cluster that autoscales, so ask for it in writing.

What does $1 per vCPU cost at your cluster size?

Managed vCPUsPer monthPer yearMarketplace tier (earlier read)
100$100$1,200Not listed
500$500$6,000$499 a month or $6,000 a year to 500 vCPUs
1,000$1,000$12,000$900 a month or $10,800 a year to 1,000 vCPUs
2,500$2,500$30,000Above the listed tiers
5,000$5,000$60,000Above the listed tiers

Up to 500 vCPUs the two routes cost the same. At 1,000 vCPUs the Marketplace tier is $1,200 a year cheaper, and it can also count toward an AWS commitment. The Marketplace tiers are a different packaging, so confirm the current listing before you rely on them. Our AWS Compute Optimizer pricing guide and the pod rightsizing pricing comparison show how the same tiers compare with the other vendors.

Is there a free trial?

Yes. The pricing file says Kubex is free for the first 60 days, with self-service signup. That is longer than most rivals: it compares with 30 days for Sedai and StormForge, and PerfectScale takes a different route with a free tier up to 300 vCPUs. Use the 60 days on your largest cluster, because a recommendation tool needs a full billing cycle of telemetry before its sizing is trustworthy.

What does Kubex replace, and what does it not?

Kubex optimizes resources. It sizes pods, nodes and instances and can apply changes through its scaler, which means it needs agent access to the cluster. It does not allocate your invoice to teams, reconcile SaaS subscriptions or explain why the bill moved month over month. Finance usually owns that second problem, and a per-vCPU tool does nothing for it. If your question is "which Kubernetes tool saves us the most", our Kubernetes rightsizing tools roundup ranks the field. If it is "where is the money going across cloud and SaaS", you need a reader of the bill instead.

QuestionKubexCostAnalyst
Price at 1,000 vCPUs$1,000 a month, $900 on Marketplace$119 a month monthly or $59 billed yearly up to $50K of tracked monthly spend
Access neededHelm-deployed agent and Prometheus dataBilling export only, read-only
OutputPod, node and instance sizing, optional automationDollar list of idle, oversized and unused cloud and SaaS spend
Best forPlatform teams that will let software resize workloadsFinance and FinOps leads who need the savings number first

They answer different questions, and many teams run both. Do the dollar review first: if the export shows that the waste is idle databases and unattached volumes, no pod scaler will touch it. The same logic runs through our ScaleOps pricing breakdown and the CAST AI pricing guide.

What to ask Kubex before you buy

  • Is a vCPU counted as provisioned, requested or averaged, and is it measured hourly or daily?
  • Do cloud VMs and RDS instances count toward the same vCPU total as Kubernetes nodes?
  • Is the trial price held for the first year, and what happens to the rate above 5,000 vCPUs?
  • Does the Marketplace tier match the direct rate, and which one draws down our AWS commitment?
  • Which actions are applied automatically, and can we keep recommendations read-only until we trust them?

Is Densify (Kubex) worth the price?

At $1 per vCPU a month the tool is cheap relative to the compute it governs. A 1,000 vCPU estate running at roughly $30 per vCPU a month in on-demand compute spends $30,000 a month, so the $1,000 fee is about 3% of that bill and pays back with a 3% reduction. The risk is not the price, it is whether your team will approve and apply the changes. Before you commit, a read-only look at your bill shows how much of the waste sits in pods and how much elsewhere. A billing export through our one-time analysis lists it in dollars. If the same team also handles month-end paperwork, Meetsetup keeps client meetings out of the inbox.

Primary sources are the Kubex pricing page and its machine-readable pricing file. To see a flat price against your own spend, see CostAnalyst's plans.

See where your cloud and SaaS money is leaking

Upload your cloud billing export and see your savings in dollars. Transparent pricing, no card to start.