Replicas vs Claude Code on the web
Claude Code on the web is Anthropic's own cloud for Claude Code: nothing to install, included in Claude plans, GitHub only, and it pushes a branch for you to turn into a pull request. Replicas runs the same Claude Code in its own isolated VM per task, on a Claude plan or an API key, opens the pull request itself, works with GitLab too, and runs Codex, Cursor and OpenCode beside it.
Checked against both vendors' docs and pricing as of November 9, 2026. This site is operated by Replicas.
Pick Replicas if
- You want to bill Claude Code to an Anthropic API key, AWS Bedrock or Microsoft Foundry, which cloud sessions don't accept.
- Your code is on GitLab, or spans GitHub and GitLab.
- You want tasks started from Slack, Linear, schedules or webhooks, and the pull request opened by the agent.
- You want Codex or other agents next to Claude Code without a second platform.
Pick Claude Code if
- Your team uses only Claude Code, your code is on GitHub, and you already pay for Claude plans: it's included and there's nothing to set up.
- You mostly pair with Claude Code in your terminal and want to push the occasional session to the cloud and back with --teleport.
Feature by feature
| Criterion | Replicas | Claude Code |
|---|---|---|
| Agents | ||
| Multi-harness support | YesClaude Code, Codex, Cursor, OpenCode, Pi and Muse Code; pick the agent per task | NoClaude Code only |
| Environment | ||
| Isolated cloud VM per task | YesEach workspace is its own isolated Linux VM (2, 4 or 8 vCPU) | YesClaude Code on the web runs each session in an Anthropic-managed Ubuntu VM (about 4 vCPU, 16 GB) |
| Workflow | ||
| Pull request output | YesThe agent opens the pull request (GitHub) or merge request (GitLab) | PartlyCloud sessions push a branch; you create the PR from the session |
| Local IDE workflow | NoAgents run only in the cloud; the desktop app and CLI connect to cloud workspaces and can open one in VS Code | YesTerminal CLI and IDE extensions; --teleport pulls a cloud session into your terminal |
| Where work starts | ||
| Start from Slack | YesMentions, DMs and threads | CheckConfirm current Slack support with Anthropic |
| Start from Linear | YesAssign or mention on an issue | CheckConfirm with Anthropic |
| Start from GitHub | YesMention on issues and PRs; can act on failed checks and reviews | YesGitHub repositories; @claude in issues and PRs through claude-code-action |
| Start from GitLab | YesIssues and merge requests; one workspace can span GitHub and GitLab | PartlyGitLab repos can be sent as a bundle, but cloud sessions can't push back |
| Models and billing | ||
| Bring your own model keys | YesAnthropic, OpenAI, AWS Bedrock, Microsoft Foundry, OpenRouter, Cursor and OpenAI-compatible keys | PartlyCloud sessions need a Claude plan; API keys and Bedrock aren't supported there (the CLI takes them) |
| Use your Claude or ChatGPT plan | YesClaude and ChatGPT subscriptions | YesPro, Max, Team and eligible Enterprise seats |
| Usage-based compute pricing | YesBilled per workspace minute by VM size; sleeping and archived workspaces cost nothing; seats are free | NoSeat-based plans from $20/mo; no separate VM charge |
| Nothing new to buy | NoA separate platform: Developer $50/mo, Team from $200/mo, each including that much usage | YesIncluded in Claude Pro, Max, Team and eligible Enterprise seats |
| Security and governance | ||
| Self-hosted or dedicated deployment | YesSelf-hosted or dedicated single-tenant deployment on Enterprise | PartlyCloud sessions are Anthropic-hosted; the CLI can run on servers you control |
| Scale | ||
| Parallel runs | YesRun many workspaces at once and compare them side by side | YesSeveral cloud sessions at once |
| Scheduled and event-driven runs | YesSchedules and GitHub, GitLab, Slack, Sentry and webhook events | PartlyThrough GitHub Actions or your own scheduler |
Sources: Replicas facts, Replicas pricing, Replicas enterprise, Replicas docs, Claude Code on the web, Claude Code GitHub Actions, Claude pricing.
Still deciding?
The comparison above is the same for everyone. The chooser weights it by your team size, the agents you use, where work starts, how you pay for models and what your security review needs, and tells you which of the two, or something else, fits.
Take the chooserCommon questions
Does Claude Code on the web work with API keys?
Anthropic's docs say cloud sessions need a Claude Pro, Max, Team or eligible Enterprise seat; API keys and Bedrock aren't supported there. Replicas runs Claude Code on either a Claude plan or an API key.
Is it the same Claude Code?
Yes. Replicas runs the Claude Code agent itself in the VM, so your CLAUDE.md, skills and MCP servers behave the way your engineers expect.
Get a second opinion on your shortlist
Tell us what you're weighing and an engineer from the Replicas team, who operate this site, will reply by email. We'll point you to whichever option fits, including when that isn't Replicas.