CodeRabbit
AI code-review bot that posts line-by-line PR feedback, summaries, and fixes on GitHub, GitLab, Azure DevOps, and Bitbucket, plus a free IDE and CLI reviewer
AI code-review bot that reviews GitHub and GitLab pull requests with full-codebase context, catching multi-file bugs before humans do
Greptile is an AI code-review bot that reads full pull requests with codebase context and comments on bugs, security risks, and style before humans review. A free Starter tier covers 50 reviews a month for one developer; the Pro plan runs $30 per seat monthly with 50 credits included and $1 per extra review. Best for teams wanting automated first-pass PR review, not feature writing.
Greptile is a Y Combinator-backed AI code-review tool that automatically reviews pull requests on GitHub and GitLab with full context of your codebase. Instead of looking at a diff in isolation, it first builds a graph index of your repository — mapping files, functions, and dependencies — then runs a parallel swarm of agents that assess how each change ripples across the project. That codebase-wide view is what lets it flag multi-file logical bugs, security risks, and style violations that a diff-only linter or a rushed human reviewer would miss.
Greptile is deliberately narrow: it reviews code, it does not write features for you. It learns your team’s conventions from past PR comments and from custom rules you write in plain English, so review feedback gets more relevant and less noisy over time. Its TREX feature goes a step further, autonomously writing and running tests for a PR in a sandbox before merge. When it finds a problem, Greptile can hand a one-click fix straight to Claude Code, Cursor, or another IDE, and its MCP integration plus the /greploop command let coding agents iterate on the fixes. The Starter tier is free for a single developer, Pro is $30 per seat per month with 50 review credits included and $1 per extra review, and Enterprise adds self-hosting in your own AWS environment.
Greptile fits teams that already have a coding assistant handling generation and now want an automated first-pass reviewer on every pull request. It shines for groups where senior reviewer time is the bottleneck, and its free Starter tier makes it easy for a solo developer or an open-source maintainer to try before committing a team to seats. Because it reviews rather than writes, it is a complement to tools like Cursor or GitHub Copilot, not a replacement for them.
Starting price: $30/mo · Free tier: yes · Model: freemium
Price history tracked from June 2026
| Plan | Price | Includes |
|---|---|---|
| Starter | Free | Unlimited repositories · 50 credits per month (1 credit = 1 standard review) · 1 active developer · Immediate access, no card required |
| Pro | $30/user/mo | 50 review credits included per seat · $1 per additional credit above 50 · Unlimited repositories and users · Custom rules written in plain English · Unlimited external app connections · 14-day free trial |
| Enterprise | Custom | Self-hosting in your own AWS environment · SSO/SAML and GitHub Enterprise support · Security and compliance features · Dedicated Slack support and custom invoicing · Custom DPA and terms |
| Pros | Cons |
|---|---|
| Genuine free Starter tier: 50 reviews a month for a solo developer with no credit card | On large diffs it can still surface review noise and false positives that reviewers have to triage |
| Full-codebase context catches multi-file bugs that diff-only linters miss | Usage-based overage: $30/seat covers 50 reviews, then $1 each — active teams averaging over 50 reviews per developer per month pay more, and the March 2026 shift to per-review billing drew criticism |
| Custom plain-English rules plus learning from PR comments cut false positives over time | Narrow scope: it reviews PRs, it does not write features for you, so you still need a coding assistant like Cursor or GitHub Copilot alongside it |
| Hands fixes directly to Claude Code and Cursor rather than only flagging problems | Requires indexing your codebase, and review quality is lower for less-common languages than for Python, TypeScript, or Go |
| Self-hosted AWS option on Enterprise for teams with strict data-residency needs |
AI code-review bot that posts line-by-line PR feedback, summaries, and fixes on GitHub, GitLab, Azure DevOps, and Bitbucket, plus a free IDE and CLI reviewer
AI code-review and governance platform that automates pull-request review and enforces standards
AI-native code editor that runs autonomous agents to build, test, and review code
AI coding assistant with inline completions, chat, and autonomous agent mode across major IDEs
Yes. The Starter tier is free and includes unlimited repositories, 50 credits per month (1 credit = 1 standard review), and 1 active developer. The Pro plan adds a 14-day free trial, and qualifying open-source projects and pre-Series A startups can get free or discounted access.
The Pro plan is $30 per seat per month, which includes 50 review credits. Additional reviews cost $1 per credit above the included 50. Enterprise pricing is custom and adds self-hosting, SSO/SAML, and dedicated support.
One credit equals one standard code review, and a TREX runtime-validation review uses 3 credits. Each Pro seat includes 50 credits per month, and extra reviews are billed at $1 per credit. In March 2026 Greptile moved from flat-rate pricing to this per-review overage model.
Yes. Greptile installs as an app on both GitHub and GitLab and reviews pull requests automatically. It also offers API access for custom products and a self-hosted deployment option in your own AWS environment on the Enterprise plan.
No. Greptile is a code-review bot: it reviews pull requests and comments on bugs, security risks, and style, but it does not write features for you. It pairs with a coding assistant such as Cursor or GitHub Copilot, which handle the actual code generation.
TREX is Greptile's runtime-validation feature. For each pull request it autonomously writes and runs tests in a sandbox environment to catch issues that static review misses. A TREX review consumes 3 credits versus 1 for a standard review.
Yes, on the Enterprise plan. Greptile can be deployed inside your own AWS environment, which suits teams with strict security, compliance, or data-residency requirements. Enterprise also adds SSO/SAML, GitHub Enterprise support, and a custom DPA.