Fleet management with agents, or a platform that builds the fleet
Plural started as enterprise Kubernetes fleet management: GitOps delivery across many clusters, Terraform run as Kubernetes resources, and pull-request automation for self-service. In 2026 it leads with AI for infrastructure. Its agents investigate alerts, plan upgrades, fix CVEs and work through tickets from inside your own clusters, and every change lands as a pull request you review. Plural’s source code is available under the AGPL or a commercial licence, and its last announced round was a $6M seed in March 2025.
Ankra overlaps with much of that. It also manages a fleet through Git, installs add-ons across clusters and runs AI with human review. The difference is where the cluster comes from and what surrounds it. Ankra builds the clusters itself and adds in-cluster CI, backups and vulnerability ranking to the same product.
If you already own a large Terraform estate on hyperscaler Kubernetes and want agents working across it, Plural fits well. If you want the clusters built for you, on European providers as well as the hyperscalers, with the platform around them, Ankra covers more.
Where the clusters come from
Plural provisions clusters by running your own infrastructure code. A pull request opens with the cluster’s name and cloud, and on merge a Terraform stack runs your module, with the plan approved by a person. Its worked example uses an AWS module, and its fleet page lists EKS, AKS, GKE, k3s, on-premises datacentres and edge sites. Each cluster runs a lightweight agent that connects outbound only, so the control plane never holds cluster-admin credentials, and clusters behind NAT catch up when they reconnect. The control plane can be Plural’s SaaS or self-hosted on a management cluster you choose. Its upgrade assistant flags deprecated APIs and links them to their place in Git.
Ankra builds kubeadm or k3s on Hetzner, OVHcloud, UpCloud, DigitalOcean, AWS EC2 and Proxmox VE, with no Terraform module for you to write, and creates EKS, GKE, AKS, DOKS, OVHcloud MKS and UpCloud UKS. It imports any conformant cluster with one Helm command, and its agent also connects outbound only. On clusters it builds, Ankra handles upgrades and node groups and power schedules, and on the five public clouds also node autoscaling and rolling OS security patching. If you prefer infrastructure as code, the Terraform provider covers everything the CLI can create.
Shipping the code and the add-ons
Plural is a strong continuous-delivery system. It syncs Helm, Kustomize and manifests across the fleet, templates Terraform outputs into Helm values, and promotes changes from development to production through pipelines with approval gates and checks between stages. Global services push baseline add-ons such as ingress and cert-manager to every cluster that matches a tag. Builds and tests stay in your own CI.
Ankra also builds the code. Applications open a setup pull request with build files, or keep the CI you already have. Ankra Pipelines run tests, rootless builds and Semgrep, Checkov and Trivy scans in your own cluster, publish only the image that passed the gate, and a merge train on GitHub tests each pull request as the commit it will merge. Stacks install Helm add-ons and manifests in dependency order, and stack profiles roll versions across a fleet. Applications start from GitHub, and some documented pipeline stage kinds do not run yet.
How each AI works
Plural’s Workbenches are configurable agents that run on demand, on a schedule or from webhooks such as Alertmanager, GitHub, Jira, Linear and ServiceNow. They combine subagents for observability, infrastructure and code changes, read a live graph of your clusters and services, and connect to other tools through MCP. Its Agent Runtime runs coding agents such as Claude Code, OpenCode and Gemini as isolated containers in your cluster, and can dispatch one to fix a detected CVE. Agents cannot push to main; every change is a pull request. You bring your own model, and Rego policies can deny tool calls or auto-approve operations. Keeping the agents inside your perimeter is a real strength for security-conscious teams.
Ankra’s AI is built into the platform. AI Insights analyses firing alerts and proposes a root cause and a fix. The AI Gateway reviews pull requests on GitHub, GitLab and Bitbucket Cloud, and Ankra AI fixes failing pull requests on their own branch. The AI board runs tickets through role agents with a person approving. Ankra’s own AI asks before every write by default, and you can bring your own model, including a self-hosted OpenAI-compatible endpoint. Its MCP server offers more than 500 tools; its writes are audited, and your MCP client decides how it confirms them.
Security, backups and cost
Plural’s security page lists SOC 2 Type 2 and GDPR. Its documentation has no backup feature guide; restore resources appear only in its API reference. Its cost management runs OpenCost in each cluster, with fleet, cluster and namespace dashboards and right-sizing recommendations. In October it released Plural Telemetry, an observability stack on object storage.
Ankra’s Security Center ranks running-workload vulnerabilities by CISA’s Known Exploited Vulnerabilities catalogue and FIRST’s EPSS scores, with optional SBOMs and compliance evidence. Backups write volumes and database dumps to a bucket you own. Cloud Cost estimates spend from list prices per namespace, stack and project, and reprices a cluster on every provider Ankra can build on. Ankra’s security and compliance material is on its trust page.
Price
Plural charges per cluster. Sandbox is free for 14 days. Pro is $450 a month for the first cluster and $250 for each additional one, up to 10, with a 30-day trial, a 99.9% uptime SLA and email support. Enterprise is priced on request, with unlimited clusters and shared, dedicated or on-premises hosting.
Ankra’s Free tier includes 30 worker vCPU. Beyond that, the Startup plan charges €0.01 per worker vCPU-hour, about €7.20 a month. Controller nodes are never billed, and your provider bills the servers.
When to choose Ankra, and when to choose Plural
Choose Ankra when you want the clusters built for you, including on Hetzner, OVHcloud, UpCloud, DigitalOcean or Proxmox VE, with applications shipped from repository to release, backups and vulnerability ranking in the same product.
Choose Plural when
- you run a large fleet on EKS, AKS or GKE and already own the Terraform that creates it
- you want agents that turn alerts and ServiceNow or Jira tickets into pull requests from inside your environment
- you need a self-hosted or on-premises management plane, or source code under the AGPL
- you need a vendor that states SOC 2 Type 2 compliance today
Compare the work that remains
- Create the foundation. Note who writes and maintains the code that creates each cluster.
- Ship a release. Include build, tests, scans and a rollback, and list every tool in the path.
- Upgrade Kubernetes. Note who finds deprecated APIs and who runs the upgrade itself.
- Restore a database. Note where backups live and how a restore is tested.
- Compare total cost. Plural charges per cluster; Ankra charges per worker vCPU after the first 30. Both sit on top of your cloud or server bill. See Plural’s pricing and Ankra’s pricing.
Reviewed 9 November 2026 against Plural’s documentation and pricing. Product scope and plans can change.