Cloud Dev EnvironmentsRebranded to OnaUpdated August 2026

Gitpod Review 2026: It's Ona Now — Pricing, Features, Pros & Cons

Item: Gitpod (Ona)
Rating: 3.9
Author: AISO Tools

Most reviews of Gitpod written this year are reviewing a product that no longer ships. Gitpod became Ona in September 2025, Gitpod Classic pay-as-you-go was switched off that October, and Ona has since announced it is joining OpenAI. Here is what actually exists today, what the $20 Core plan buys, and — if you came here looking for the old thing — where to go instead.

⚠️ Read this before you evaluate

If you followed an "Open in Gitpod" badge, a README link or a 2024 tutorial, the product it describes is gone. Gitpod Classic PAYG sunset on October 15, 2025. The company now trades as Ona and gitpod.io serves Ona's pricing. Ona has announced it is joining OpenAI, which is a real consideration for anyone about to standardise a whole engineering team on it.

Quick Verdict

3.9/5

Overall Rating

$20/mo

Core, up to 100 seats

OCU credits

Expire monthly

Best for: teams that want to run coding agents in isolated, reproducible cloud environments and like that seats are free. Not for: anyone who specifically wanted Gitpod Classic, or a team that needs roadmap stability from an independent vendor.

What Happened to Gitpod — The Timeline

When	What changed	Why it matters to a buyer
September 2025	Gitpod rebrands to Ona	Same team, new name, repositioned as 'mission control for software engineering agents'
October 15, 2025	Gitpod Classic PAYG sunsets	The pay-as-you-go product most old tutorials describe is switched off
2026	Ona announces it is joining OpenAI	Announced on ona.com — roadmap and pricing independence now sit with an acquirer
Today	gitpod.io redirects into Ona	The Gitpod pricing URL serves Ona's Core/Enterprise plans

What Is Gitpod (Ona) Now?

Gitpod's original pitch was one of the cleanest in developer tooling: describe your development environment in the repository, and every contributor gets an identical, disposable workspace on demand. No "works on my machine", no two-day onboarding, no laptop rebuild after a bad dependency upgrade. That capability is still present in Ona.

What changed is the job the product is sold for. Ona positions itself as mission control for software engineering agents: the isolated, reproducible environment is now primarily a safe sandbox for an agent to work in, with a human reviewing what came out. Unlimited parallel environments matters much more in that framing — you are running a fleet, not a session.

Practically, Core gives you environments up to 32 cores, 128 GB RAM and 200 GB disk with GPU support, prebuilds to cut cold starts, 50 automations with up to 10 parallel actions, role-based access management, org-wide command and deny lists, and MCP server support. Enterprise adds the deployment and governance layer: self-hosted VPC, network control, SSO/OIDC, org-wide secrets, warm pools, audit trails on every human and AI action, API/SDK access and SLAs.

Gitpod / Ona Pros & Cons

✓ Pros

•The core promise still holds: a reproducible development environment defined in the repo, started on demand, so onboarding a new engineer is a URL rather than a two-day laptop-setup ritual
•Core is not priced per seat — up to 100 team members are included at $20/month, with credits pooled across the organisation, which is unusual in a category where per-developer billing is the norm
•Unlimited parallel environments on both tiers means agent runs and human sessions do not compete for a fixed pool of slots
•Environment specs go up to 32 cores, 128 GB RAM and 200 GB disk, with GPU support — enough to run workloads that a laptop cannot
•You can connect an existing Codex subscription on Core, so model usage is covered by a plan you already pay for and Ona bills only for compute and environments
•Claude Code can be installed and authenticated inside an environment, with Anthropic billing the model usage separately — the platform is not locked to a single agent vendor
•Prebuilds and (on Enterprise) warm pools address the single biggest complaint about cloud dev environments, which is the cold-start wait before you can type
•Enterprise is a genuinely serious deployment story: self-hosted Ona-managed VPC, full network control, SSO/OIDC, org-wide secrets and audit trails on every human and AI action

✗ Cons

•The name you searched for no longer exists as a product. Gitpod rebranded to Ona in September 2025, and gitpod.io now serves ona.com — if you were evaluating 'Gitpod', you are evaluating something with a different strategic centre of gravity
•Gitpod Classic pay-as-you-go sunset on October 15, 2025. If you are arriving from an old tutorial, blog post or README badge, the thing it describes is switched off
•Ona has announced it is joining OpenAI. Whatever you think of that outcome for the team, it is a material fact for a buyer: roadmap, pricing and independence are all now downstream of an acquirer's priorities
•The product's centre moved from 'cloud dev environments for humans' to 'mission control for software engineering agents'. If you wanted the former, you are now buying it as a component of the latter
•OCU credit billing is hard to forecast. Ona's own examples range from 1 OCU to explain a small codebase to 8 OCUs to add a feature to a medium one, and the docs explicitly warn of large variability per task
•Monthly subscription OCUs expire at the end of the billing month — unused credits do not roll over, so light months are effectively forfeited spend
•Environments auto-delete after 7 days of inactivity on Core, which is fine for active work and annoying for the branch you come back to next sprint
•Almost everything a compliance-bound team needs — SSO and OIDC, audit logs, org-wide secrets, MCP usage controls, API and SDK access, SLAs — is Enterprise-only, and Enterprise has no published price

Gitpod / Ona Pricing 2026

Core

from $20/mo

•80–2,200 OCUs included monthly
•Up to 100 team members, no per-seat fee
•Unlimited parallel environments
•Up to 32 cores / 128 GB RAM / 200 GB disk, GPU support
•Prebuilds, 50 automations, basic support

Individuals and teams who want reproducible environments without a procurement cycle

Best for Teams

Enterprise

Contact sales

•Self-hosted, Ona-managed VPC deployment
•SSO/OIDC, org-wide secrets, audit trails
•Warm pools for instant environment starts
•Programmatic access via API and SDK
•Uptime SLAs and dedicated support

Regulated engineering orgs that need network control and audit evidence

Add-on OCUs start at $10 per 40 OCUs. Monthly subscription credits expire at the end of the billing month; add-on credits are valid for one year. Plan contents as published on Ona's pricing page in August 2026 — verify before purchase.

Who Should Actually Use It

Use it if: you are actively building an agent-assisted engineering workflow and want the agents running somewhere isolated and reproducible rather than on a developer's laptop; you have a team large enough that free seats up to 100 people is a real saving against per-developer competitors; or you already pay for Codex and want to reuse that subscription while paying only for compute.

Skip it if: what you wanted was Gitpod Classic, in which case Codespaces, Coder or DevPod are the honest recommendations; if unpredictable credit consumption is worse for you than a slightly higher fixed bill; or if you are making a multi-year platform commitment and need the vendor's roadmap to be its own, which — with the OpenAI move announced — it no longer entirely is.

Frequently Asked Questions

Is Gitpod still available in 2026?

Not under that name, and not in the form most people remember. Gitpod rebranded to Ona in September 2025, and the gitpod.io pricing page now serves Ona's plans. Gitpod Classic — the pay-as-you-go browser-based workspace product that the majority of tutorials, README badges and 'Open in Gitpod' buttons point at — sunset on October 15, 2025. The open-source repository is still on GitHub under gitpod-io/gitpod with a notice to that effect. So if you clicked a Gitpod link and landed here: the company exists, the team exists, the specific product you were looking for does not.

What is Ona and how is it different from Gitpod?

Ona is the same company with a substantially different thesis. Gitpod sold cloud development environments to humans: define the environment in the repo, get an identical workspace on demand, stop debugging someone's laptop. Ona sells that capability as the substrate for autonomous software engineering agents — the environments are still there, but the framing is a fleet of agents doing work in isolated sandboxes with a human supervising. If you want the original value proposition, you can still get it; you are just buying it inside a product optimised for a different job.

How does Ona pricing work in 2026?

Core starts at $20/month and includes between 80 and 2,200 recurring monthly OCUs depending on the plan level you pick, with add-on credits from $10 per 40 OCUs. OCUs — Ona Compute Units — are a single normalised meter covering both agent token usage and the infrastructure your environments consume while running. Seats are not billed: up to 100 team members can join a Core organisation and the credit pool is shared across all of them. Monthly subscription credits expire at the end of the billing month; purchased add-on credits are valid for a year. Enterprise pricing is not published.

How many OCUs will I actually burn?

Ona publishes examples and then explicitly warns that variability is large. Their figures: explaining a small codebase is about 1 OCU, running a regular 4 vCPU / 16 GB VM for an hour is about 1 OCU, explaining a large complex codebase is around 3, creating a new web app from scratch around 4, fixing a bug in a large complex codebase around 5, running a GPU VM (16 vCPU / 64 GB) for an hour around 7, and adding a feature to a medium codebase around 8. The practical read: an $20 Core plan with 80 OCUs is roughly ten meaningful agent tasks or eighty VM-hours, and a team leaning on agents will be buying add-on credits. Model your own first month before assuming the base plan covers you.

Can I use Claude Code or Codex with Ona?

Both, and this is one of the better things about the product. On Core you can connect an existing Codex subscription — your subscription covers model usage and Ona bills only for environments and compute, which avoids paying twice for inference. Claude Code can be installed and authenticated inside an Ona environment, with Anthropic billing you directly for the model. That decoupling of compute from inference is worth more than it sounds: it means the OCU meter is measuring infrastructure you can reason about rather than an opaque bundled markup on tokens.

What should I use instead if I just want cloud dev environments?

GitHub Codespaces is the obvious default if your code is already on GitHub — it is the closest like-for-like replacement for Gitpod Classic and the per-hour pricing is legible. Coder is the answer if you need self-hosted environments in your own infrastructure without an Enterprise sales conversation. DevPod is the open-source, client-side option that runs the same devcontainer spec against any backend including your own machine, which is the least lock-in available. And if the actual requirement is a fast throwaway sandbox rather than a full development environment, StackBlitz or Replit will get you there in less time than configuring any of the above.