Eclipse Che Alternatives for Cloud Development Environments

Eclipse Che is an open source platform that runs developer workspaces as pods on Kubernetes or OpenShift. Each workspace is defined by a devfile, and you open it in a browser-based VS Code or connect IntelliJ IDEA through JetBrains Gateway, per the Che documentation. The project is very much alive. Its releases page shows a new version almost every month, from 7.119.0 on June 19 to 7.124.0 in October 2026.

The hard part is everything around it. Che is installed through an Operator, so somebody has to own a cluster, its identity provider and its storage before the first workspace starts. Teams look for alternatives when that cost outweighs the benefit, when developers want their local IDE instead of a browser tab, or when they standardise on devcontainer.json rather than devfiles. The six picks cover self-hosted, hosted and local options.

Prices below are as of November 8, 2026. Our method is on How we review tools.

Quick comparison

Coder

Coder keeps what Che users like, self-hosting and environments defined as code, and drops the requirement that workspaces live as Kubernetes pods. Templates written in Terraform can provision a pod, a cloud VM or a Docker container, and developers connect from VS Code, JetBrains IDEs, Jupyter or a browser.

The Coder pricing page makes the Community edition free for unlimited self-hosted workspaces, with up to five active Coder Agents for AI tools such as Claude Code, Codex or Cursor and one platform integration. Premium adds audit logs, high availability, workspace proxies for distant offices, multi-org RBAC and SLA support, billed annually per user after a demo.

Best for: Platform teams that want Che-style control but need workspaces on VMs as well as Kubernetes.

Pricing:

• Community: free, unlimited self-hosted workspaces
• Premium: price on request, billed annually per user
• AI Premium: custom pricing, unlimited Coder Agents

Watch out for:

• The Premium price is not published, so budgeting for audit logs or HA means a sales call.
• You still run and patch the control plane yourself; Coder offers no managed hosting on its pricing page.

Official site: Coder.

GitHub Codespaces

Codespaces is the easiest way to stop running Che yourself if your code is already on GitHub. Each repo can carry a devcontainer.json, and a codespace opens in the browser or in desktop VS Code. JetBrains IDEs and JupyterLab are supported in beta.

Billing is per hour. The Codespaces billing docs list $0.18 an hour for 2 cores up to $2.88 for 32 cores, plus $0.07 per GB-month of storage. GitHub Free accounts get 120 core hours and 15 GB-month included, and GitHub Pro gets 180 core hours and 20 GB-month.

Best for: Teams on GitHub that want hosted environments with no cluster to run.

Pricing:

• Free personal accounts: 120 core hours and 15 GB-month storage a month
• GitHub Pro: 180 core hours and 20 GB-month
• Compute: $0.18 (2-core), $0.36 (4-core), $0.72 (8-core), $1.44 (16-core), $2.88 (32-core) per hour
• Storage: $0.07 per GB-month

Watch out for:

• Workspaces run on GitHub's cloud, so code and secrets leave your own network, which is often the reason a company chose Che.
• Core hours go fast: the free 120 hours equal only 60 hours on the smallest 2-core machine.

Official site: GitHub Codespaces.

Google Cloud Workstations

Cloud Workstations is Google's managed take on the Che idea. Admins define workstation configurations from container images, and developers get a VM-backed environment inside the company's own Google Cloud project and VPC, opened in a browser editor or a local IDE.

Costs stack in layers on the Cloud Workstations pricing page: a management fee of $0.05 per vCPU per hour while a workstation runs, $0.20 per cluster per hour for the control plane, the VM itself and persistent disk. An e2-standard-4 in us-central1 comes to about $0.334 an hour for compute plus the management fee, plus $0.20 an hour per cluster and disk.

Best for: Companies already on Google Cloud that want managed workstations inside their own VPC.

Pricing:

• Management fee: $0.05 per vCPU per hour while running
• Control plane: $0.20 per cluster per hour, always on
• Example: e2-standard-4 about $0.334 an hour in us-central1, USD
• Persistent disk: billed separately, $0.10 per GB-month in Google's example

Watch out for:

• The control plane fee runs around the clock, about $146 a month per cluster, even with no workstation in use.
• It ties your development environments to one cloud provider.

Official site: Google Cloud Workstations.

VS Code Dev Containers

Many teams do not need a remote platform at all. The Dev Containers extension opens a project inside a container on the developer's own machine or a remote Docker host, with extensions running in the container and the environment described in .devcontainer/devcontainer.json. The same file works in Codespaces, so you can start local and move to hosted later.

It needs Docker Desktop on Windows or macOS, or Docker CE with Compose on Linux, per the Dev Containers docs. Remote hosts need at least 1 GB of RAM. If VS Code itself is the problem, our VS Code alternatives page lists other editors.

Best for: Small teams that want reproducible environments without running any server.

Pricing:

• Free: VS Code and the extension cost nothing
• Docker Desktop at larger companies needs a paid plan, from $9 a month billed yearly

Watch out for:

• Windows container images are not supported, and other Docker-compatible CLIs such as Podman are not officially supported.
• Bind-mounted files are slower on Windows and macOS, which hurts big repos.

Official site: VS Code Dev Containers. Our listing: VS Code Dev Containers.

Red Hat OpenShift Dev Spaces

Some teams leaving Che really want someone to call when it breaks. OpenShift Dev Spaces is Red Hat's supported product built on the Che code, installed from OperatorHub, with admins curating the extensions developers may install from an internal registry.

The Dev Spaces overview lists VS Code, IntelliJ, Cursor and Amazon Kiro as supported IDEs and says the product is included with a Red Hat OpenShift subscription. A no-cost trial runs in the Developer Sandbox.

Best for: Enterprises already paying for OpenShift that want Che with vendor support.

Pricing:

• Included with a Red Hat OpenShift subscription; no separate price
• Free trial through the Developer Sandbox

Watch out for:

• It is still Che underneath, so the cluster overhead that pushes people away stays.
• It only runs on OpenShift, not on plain Kubernetes distributions.

Official site: Red Hat OpenShift Dev Spaces.

Ona (formerly Gitpod)

Gitpod was the best-known Che rival for years. The German company renamed itself Ona in September 2025 and rebuilt the product around AI agents that run in cloud environments, and in June 2026 OpenAI announced it would acquire Ona to run Codex agents inside customers' own clouds, as The Next Web reported.

Ona's site now describes it as part of OpenAI, and sign-in still goes through app.gitpod.io. Environments run in your own cloud account, which keeps Che's main selling point of code staying inside your network.

Best for: Teams that want cloud environments mainly to run AI coding agents for long tasks.

Pricing:

• Price on request; the pricing page is a contact form

Watch out for:

• The future of the standalone product is unclear. The deal folds Ona's platform into Codex, and no public plan for existing customers has been announced.
• Without public prices, it is hard to compare against Codespaces or Coder Community.

Official site: Ona (formerly Gitpod).

How to choose

When to stay with Eclipse Che

Stay on Che if you already run Kubernetes well and want a free, open source platform where workspaces never leave your cluster. It handles OIDC sign-in, Kubernetes RBAC and Prometheus monitoring out of the box, per the Che overview, and monthly releases keep it current. Coder Community and Che are the only free self-hosted choices here.

Frequently asked questions