CodeRabbit

What it does

Writes code

PartlyCode authoring lives in the separately billed Agent and Cloud Coding Agent add-ons, not in the seat price.you can ask CodeRabbit to investigate your codebase, generate an implementation plan, and open a pull request — without switching to another tooldocs.coderabbit.ai/agent ↗

Debugs and fixes

PartlyFix CI failures investigates failing checks and opens a stacked PR with the fix; Team plan and above, GitHub and Azure DevOps only.When CI fails on a PR, you can ask CodeRabbit to investigate the failing checks and fix them for you.docs.coderabbit.ai/finishing-touches/fi… ↗

Writes tests

PartlyUnit test generation is a Team-plan finishing touch triggered by a PR comment or walkthrough checkbox; open beta.Type @coderabbitai generate unit tests in a GitHub, GitLab, or Azure DevOps pull request or merge request comment to generate tests automaticallydocs.coderabbit.ai/finishing-touches/un… ↗

Reviews code

YesReviewing someone else's diff is the core product: automatic line-by-line review on every pull request.Within moments of opening a pull request, CodeRabbit analyzes your code with multiple AI models and provides actionable feedbackdocs.coderabbit.ai/overview/pull-reques… ↗

Whole-codebase work

PartlyReviews reason across linked repositories (1 repo on Essentials up to 20 on Enterprise), but it reviews and patches changes rather than driving a repo-wide migration.Link related repositories so CodeRabbit can detect breaking changes, API mismatches, and dependency issues that cross repository boundaries during…docs.coderabbit.ai/knowledge-base/multi… ↗

Ships it

NoPost-merge automation opens follow-up PRs, posts reports and files tickets; nothing in the published action set deploys.It can open a follow-up pull request, post a report on the PR, or deliver output to a connected integration.docs.coderabbit.ai/pr-reviews/post-merg… ↗

How much it does unattended

Runs autonomously

YesReviews run unattended on every eligible PR by default; code-changing actions such as Autofix or unit tests need an explicit command or checkbox.By default, enabled is true and CodeRabbit reviews every eligible PR automaticallydocs.coderabbit.ai/configuration/auto-r… ↗

Multi-agent

YesEach review is orchestrated across parallel specialised agents rather than one prompt.Specialized AI agents working in parallel: Review, Verification, Chat, Pre-Merge Checksdocs.coderabbit.ai/overview/architecture ↗

Agent permissions

PartlyCommentary is automatic; every code-writing action is opt-in per PR, and Slack Agent access is bounded by scopes over repositories, connections and spend.Every action is previewed as a checkbox in the PR walkthrough before merge, so authors and reviewers can skip actions that do not fit.docs.coderabbit.ai/pr-reviews/post-merg… ↗

Which models it runs on

Claude

YesAnthropic is one of the two inference providers CodeRabbit sends review context to; the model is chosen by CodeRabbit, not by you.Your code is shared with OpenAI and/or Anthropic for reviewing purposes onlydocs.coderabbit.ai/faq ↗

GPT

YesOpenAI is the other named inference provider for hosted reviews.Your code is shared with OpenAI and/or Anthropic for reviewing purposes onlydocs.coderabbit.ai/faq ↗

Open models

NoThe hosted service names only OpenAI and Anthropic as inference providers; no open-weight option is published for CodeRabbit Cloud.Your code is shared with OpenAI and/or Anthropic for reviewing purposes onlydocs.coderabbit.ai/faq ↗

Model choice

PartlyOnly self-hosted Enterprise deployments choose the model provider; on the hosted service CodeRabbit picks the models.Bring your own model — connect CodeRabbit to your own large language model provider or account.docs.coderabbit.ai/self-hosted/overview ↗

Where you use it

In your editor

YesExtension for VS Code and VS Code-based editors reviews uncommitted changes; free tier included.Get instant feedback in VS Code, Cursor, Windsurf, and any VS Code-based IDE.docs.coderabbit.ai/overview/ide-cli-rev… ↗

On the command line

YesTerminal reviewer for macOS, Linux and Windows that other coding agents can call.Smart CLI Reviews act as quality gates for Codex, Claude, Gemini, and you.coderabbit.ai/cli ↗

In your pipeline

YesReviews post on pull requests across GitHub, GitLab, Bitbucket and Azure DevOps, and the CLI runs inside your own pipeline.Run CodeRabbit review before you open a pull request or as a part of your custom CI/CD pipeline.docs.coderabbit.ai/overview/ide-cli-rev… ↗

In a browser

Yesapp.coderabbit.ai carries the dashboard, Triage queue, Change Stack review interface and settings.The CodeRabbit Dashboard provides visibility into your team's review speed, code quality, collaboration patterns, and the impact of AI-assisted…docs.coderabbit.ai/guides/dashboard ↗

Whose machine it runs on

Self-hosted

PartlySelf-hosting is Enterprise-only and gated on 500+ seats; it ships as a container image you run yourself.The self-hosted option is available for CodeRabbit Enterprise customers with 500 or more user seats.docs.coderabbit.ai/self-hosted/overview ↗

Not established: Open source

What it costs to run

How it meters

YesPer-developer seats, charged only for developers who open PRs, plus metered extras: $0.25 per reviewed file over the limit and $0.50 per agent minute.$24 per developer per month billed annually, or $30 month-to-month.docs.coderabbit.ai/management/plans ↗

Free tier

YesFree plan needs no card and no call; open-source repositories get Team features free, and new organisations start on a 14-day Advanced trial.Unlimited public and private repositories, no credit card required.docs.coderabbit.ai/management/plans ↗

API access

PartlyREST API at api.coderabbit.ai covers metrics, learnings, users, roles and audit logs, but the plan badge and pricing page put it on Enterprise.The CodeRabbit API provides programmatic access to review data, AI Deep Scan findings, and administrative operations.docs.coderabbit.ai/api ↗

MCP server

PartlyCodeRabbit connects out to MCP servers for review context (5 connections on Essentials to 20 on Enterprise); other agents drive it through installed skills and the CLI's agent output mode, and no CodeRabbit MCP server is documented.The CodeRabbit CLI detects and installs user-level skills for Codex, Claude Code, Cursor, Gemini CLI, and GitHub Copilot.docs.coderabbit.ai/cli/skills ↗

Bring your own key

PartlyYour own LLM provider or account is possible only on self-hosted Enterprise; hosted reviews give no key or cloud choice.The agent sends review prompts, source code, and related data to your configured large language model provider for inference.docs.coderabbit.ai/self-hosted/overview ↗

Buying it for a team

A company can buy it

YesOrganisations buy seats against their Git organisation; only developers who open pull requests consume one.No, you will only be charged for the developers who create pull requests. You also have the option to manually assign seats.coderabbit.ai/faq ↗

Seat model

YesPer-developer seats, reassignable at any time, with automatic or manual assignment; no published minimum except the 500-seat floor for self-hosted Enterprise.You can purchase any number of seats and assign them to specific developers.coderabbit.ai/faq ↗

Pooled budget

PartlyReview allowances are per developer per hour, while pay-as-you-go spend is one org-wide path with a single monthly cap.There is one pay-as-you-go billing path covering all PR and CLI reviews.docs.coderabbit.ai/management/usage-bas… ↗

Admin controls

YesAdmins set organisation-wide review configuration and global overrides that outrank repository settings; custom RBAC roles are Enterprise-only.You can use the CodeRabbit web interface to set the CodeRabbit configuration for all Git repositories associated with your organization.docs.coderabbit.ai/guides/organization-… ↗

Audit log

PartlyAdministrative audit log exists but is restricted to Enterprise organisations and admin roles.The Audit Logs give Enterprise organizations a searchable record of administrative actions taken across workspace settings, billing controls, seat…docs.coderabbit.ai/management/audit-logs ↗

Single sign-on

PartlySSO workspaces with SCIM provisioning are an Enterprise line item; the docs reference SAML attributes but publish no explicit SAML or OIDC support list.Includes everything in Advanced, plus self-hosting options, multi-organization support, SSO, SLA support with a dedicated Customer Success Managerdocs.coderabbit.ai/management/plans ↗

What happens to your code

Opt out of training

YesNo training on customer code by CodeRabbit or its providers; retention of learnings and review context can be switched off, and self-hosted can disable retention entirely.CodeRabbit never uses customer code for trainingdocs.coderabbit.ai/faq ↗

Data residency

PartlyUS region by default with an EU-hosted account option; EU SaaS deployment is listed as an Enterprise item, and self-hosting keeps code inside your own network.For a first login to an EU-hosted CodeRabbit account, use coderabbit auth login --region eu. When no region is specified or saved, the CLI uses the…docs.coderabbit.ai/cli/headless-cli-int… ↗

Getting out

PartlyCSV export of per-PR metrics and Learnings is available from the dashboard, and the Metrics Data API on Enterprise; usage is billed after the fact rather than as prepaid credits, and no expiry policy is published.The Data Export page provides CSV downloads of per-PR review metrics for offline analysis, reporting, or integration with other tools.docs.coderabbit.ai/guides/dashboard-met… ↗

Certifications

YesSOC 2 Type II and GDPR are claimed; no ISO 27001 or HIPAA claim appears on the site or in the documentation.CodeRabbit is SOC 2 Type II certified and GDPR compliant, ensuring that your data remains confidential and is used solely to fine-tune your reviews.coderabbit.ai/faq ↗