Skip to main content

AI Code Review

Qodo

Qodo reviews pull requests on GitHub, GitLab, Bitbucket, Azure DevOps, and Gerrit against your rules and repository context. Qodo Merge was the former name.

The 30-Second Take

Pick Qodo if one reviewer must enforce written standards across repositories on any of five Git platforms.

Qodo Review, formerly Qodo Merge, checks each pull request for bugs, rule violations, requirement gaps and cross-repository breaking changes. It runs on GitHub, GitLab, Bitbucket, Azure DevOps and Gerrit.

Best fit
Pull request review across several repositories and Git platforms, enforcing written team rules.
Main constraint
After the 14-day trial, every review draws from a paid credit pack; only qualified open-source projects get free access.
First move
A trial on one busy repository, with Review Standards written and an overage cap set before the trial ends.

Why

Watch the cost per review: about 18 reviews fit in a US$30 credit pack, by Qodo's own estimate, and every review is metered once the trial ends. Its Agentic Toolbox runs the same review inside coding agents before a pull request exists.

Practitioner sentiment, leaders and teams+17 Mixed

23 practitioners, 68 sources reviewed, Updated Oct 25, 2026How this is measured

Read the GILD Review
Save a product your organization uses.Keep up with Qodo without adding it to your stack.Visit Website
Operating stack
Agentic Software Development

Comparisons

How does Qodo compare with the alternatives?

  1. Qodo vs CodeRabbit

    • AI Code Review

    Choose Qodo for one credit pool per workspace or on-premises below CodeRabbit's 500-seat minimum, and CodeRabbit for a free tier and an hourly allowance per developer. Price one month of your own pull requests on each before you pick.

  2. Qodo vs GitHub Copilot

    • AI Code Review

    Start with the code review in Copilot Business or Enterprise if you work on GitHub.com. Add Qodo when repositories also sit on GitLab, Bitbucket, Azure DevOps, or Gerrit, or when rules must hold across repositories.

  3. Qodo vs Greptile

    • AI Code Review

    Both bill reviews from one shared credit pool with no seat fees. Choose Qodo for a monthly credit pack or code on Azure DevOps or Gerrit, and Greptile for prepaid US$1 credits with no plan fee. Both read beyond the diff and run on your own infrastructure on Enterprise plans.

Operator Insights

What are engineering leaders learning from deploying Qodo?

See all 3 operator accounts

Buying and deployment

What engineering leaders should verify before a broader rollout.

Sourced from Qodo documentation

How you buy it

US$30

A month for Pro Team with 2,500 credits. The card says 30 users; the FAQ sets no user limit.

Source 1, qodo.ai, Pricing: PricingSource 2, Qodo, Pricing & usage: Pricing & usageSource 5, Qodo, Qodo for Open Source Projects: Qodo for Open Source Projects

Details

Qodo offers a 14-day trial with unlimited reviews and no card. Pro Team starts at US$30 a month for 2,500 credits at US$0.012 each, billed monthly. Pro Team has no user limit, per Qodo's FAQ and billing docs, and the 30-user figure on its pricing card is sizing guidance. Enterprise pricing is custom. Public GitHub repositories with 200 or more stars can get free open source reviews.

Cost drivers

About US$1.67

A review, by GILD's math from Qodo's estimate that the Pro Team credits cover 18 reviews.

Source 1, qodo.ai, Pricing: PricingSource 2, Qodo, Pricing & usage: Pricing & usage

Details

Reviews draw from one workspace credit pool. Larger pull requests use more credits, and credits expire monthly. Qodo estimates about 18 reviews per 2,500 credits, or about US$1.67 a review by GILD's calculation. Overage costs the same per credit, up to the customer's cap.

Where it runs

Five Git platforms

GitHub, GitLab, Bitbucket, Azure DevOps, and Gerrit. Qodo set October 30, 2026 to retire its IDE plugin.

Source 3, Qodo, Deployment models by Git provider: Deployment models by Git providerSource 4, Qodo, Agentic Toolbox overview: Agentic Toolbox overviewSource 6, Qodo, Integrations (IDE plugin retirement): Integrations (IDE plugin retirement)

Details

Pull request review runs on GitHub, GitLab, Bitbucket Cloud and Data Center, Azure DevOps, and Gerrit. The Agentic Toolbox also runs local reviews in Claude Code, Codex, Kiro, and MCP clients. Qodo set October 30, 2026 as the IDE plugin's retirement date and points local review to the Agentic Toolbox.

What to verify

No training on code

Qodo says it does not train on your code. Air-gapped, SSO, and audit logs need Enterprise.

Source 1, qodo.ai, Pricing: PricingSource 3, Qodo, Deployment models by Git provider: Deployment models by Git providerSource 7, Qodo, Review Standards: Review StandardsSource 8, Qodo, Relevance: Relevance

Details

Qodo states that it does not train models on your code. Gerrit support and single-tenant, on-premises, and air-gapped deployments require Enterprise, as do BYOK, SSO/SAML, audit logs, and a DPA. On-premises deployment excludes Azure DevOps Cloud. Review Standards and Relevance are in beta.

Signals

What is the current evidence saying about Qodo?

A current read of public reports from engineering leaders, teams, maintainers, and individual developers about Qodo, Qodo Merge, and PR-Agent. Every finding links to its sources.

Whose sources count

Practitioner sentiment, leaders and teams

+17 Mixed net

of 23 sources that take a side, each author counted once. Small sample.

How this figure is made

Current sentiment

Positive, mixed and negative sources

Of 23 team and maintainer accounts that give an opinion, 9 speak well of Qodo, 9 report both gains and problems, and 5 speak against it. All 7 team sources that discuss bug catches report real ones, and comment noise is raised by 8 of the 10 that discuss it. At Beko, developers resolved 73.8% of the reviewer's comments while average pull request closure time rose from 5 hours 52 minutes to 8 hours 20 minutes. Of the 8 teams that weighed keeping Qodo, 5 kept it and 3 dropped it.

Run a pilot on live pull requests to measure which comments engineers act on and which they dismiss.

Evidence field. All 68 sources, one mark each.

Arrange by

Each mark is one source: positive, mixed, negative, no opinion stated, or documentation (not counted).

By sentiment, Engineering leaders and teams. Every source GILD read, grouped as the field draws it.
GroupSourcesPositiveMixedNegativeNo opinion stated and not counted in the scoreDocumentation, checked and not counted
Positive990000
Mixed909000
Negative500500
Not counted100010
Documentation15000015
Outside this group291112600
Every source GILD read, by the quarter it was published, as the field's time axis counts it. Documentation is dated by the day it was checked.
QuarterSources
Q1 20251
Q2 20252
Q3 20252
Q4 20256
Q1 20267
Q2 20265
Q3 202626
Q4 20264
Documentation, checked15

By sentiment

15 documentation pages check facts and are not counted; 29 sit outside this group.

Hover a mark for a quick look, click it or press Enter to open its source.Tap a mark to open its source, then step to its neighbors with Previous and Next.

What practitioners praise

Top praise of 2. Each mark is a source on it; filled marks praise it.

  1. Real bug catchesAll 7 sources on it praise it

Most common concerns

Top concern of 2. Each mark is a source on it; filled marks raise it.

  1. Comment noise and wrong findings8 of 10 sources on it raise it as a concern
All 68 sources behind these findings
  1. ci(review): pilot Qodo-first review with bounded rounds (vox-2q95)Jim Freeman, founder at Punt Labs (Punt Labs, open source tools for agentic software engineering teams, with AI agents opening most pull requests), GitHub pull request on punt-labs/vox, published Oct 23, 2026
  2. [build] automatically generate Qodo reviews from PRs by committersTitus Fortner, Technical Leadership Committee member at Selenium (Selenium browser automation project, multi-language monorepo with a committers team and outside contributors), GitHub pull request on SeleniumHQ/selenium, published Oct 12, 2026
  3. Qodo pricing & usageQodo (Official), Product documentation, checked Oct 12, 2026
  4. Qodo for Open Source ProjectsQodo (Official), Product documentation, checked Oct 12, 2026
  5. Deployment modelsQodo (Official), Product documentation, checked Oct 12, 2026
  6. Deployment models by Git providerQodo (Official), Product documentation, checked Oct 12, 2026
  7. How Qodo code review worksQodo (Official), Product documentation, checked Oct 12, 2026
  8. Severity thresholdsQodo (Official), Product documentation, checked Oct 12, 2026
  9. RelevanceQodo (Official), Product documentation, checked Oct 12, 2026
  10. Configuration fileQodo (Official), Product documentation, checked Oct 12, 2026
  11. Review StandardsQodo (Official), Product documentation, checked Oct 12, 2026
  12. Qodo code review analyticsQodo (Official), Product documentation, checked Oct 12, 2026
  13. Git provider feature supportQodo (Official), Product documentation, checked Oct 12, 2026
  14. Platform architectureQodo (Official), Product documentation, checked Oct 12, 2026
  15. Sunsetting legacy v1 commandsQodo (Official), Product documentation, checked Oct 12, 2026
  16. Agentic Toolbox overviewQodo (Official), Product documentation, checked Oct 12, 2026
  17. IntegrationsQodo (Official), Product documentation, checked Oct 12, 2026
  18. Docs policy: generated site output is committed, not hand-edited - Qodo rule 1255924 conflicts with .pr_agent.toml ignore globmyakove, mtv-api-tests collaborator (self-described: Red Hat QE, Migration Toolkit for Virtualization API test suite), GitHub issue, published Oct 8, 2026
  19. Lower the advisory quorum to 5: drop the three reviewers that produce no countable signal (re-file of #1894 AC #3)don-petry, fleet operator for the petry-projects organization (self-described: petry-projects repositories with an automated review quorum), GitHub issue, published Oct 1, 2026
  20. task: qodo-pr-agent - Validate pilot credential boundary and activationChris Vancoillie (eleshar), LightSpeed WordPress Development (self-described: LightSpeed (lightspeedwp), an organization-wide pilot of self-hosted PR-Agent with an Anthropic key capped at US$20 a month), GitHub issue, published Sep 24, 2026
  21. Great Cross-Repo PR Insights, Though Severity Flags Can Miss the MarkRahul S., reviewer on G2 (self-described: Team runs Qodo on all pull requests in its AI development lifecycle), G2 review, syndicated on AWS Marketplace, published Sep 24, 2026
  22. Compare Qodo and Greptile reviews and pick oneViktor Farcic (vfarcic), repository owner (self-described: dot-agent-deck open source repository), GitHub issue, published Sep 23, 2026
  23. Run PR-Agent from CI instead of the metered Qodo reviewVictor Lyuboslavsky (getvictor), Victor Lyuboslavsky, Fleet Device Management (RawTrail, formerly fleet-edr) (self-described: fleet-edr, a public repository that merged 227 pull requests in 30 days with Qodo reviewing every push), GitHub issue, published Sep 21, 2026
  24. chore: suppress empty Qodo review commentsFooftilly, repository owner (self-described: PRKS repository), GitHub pull request, published Sep 20, 2026
  25. review config: Qodo Rule 232249 contradicts CLAUDE.md's atomicity rule (and its own cited precedent #1028)cptkoolbeenz, DAQiFi firmware team member (self-described: DAQiFi Nyquist firmware (FreeRTOS on PIC32)), GitHub issue, published Sep 18, 2026
  26. Evaluate a replacement for qodo PR review (trial ended)blehnen, Maintainer of DotNetWorkQueue (self-described: DotNetWorkQueue, an open source .NET library below Qodo's 200-star open source threshold), GitHub issue, published Sep 17, 2026
  27. fix(.pr_agent.toml): contain Qodo skill-insight/cross-repo leak of private-repo contentekson73, repository owner (self-described: multi-agent-os repository with a private downstream repository), GitHub pull request, published Sep 14, 2026
  28. Qodo review bot has failed on every recent PR since Aug 29auroraxo, PR-Agent contributor (self-described: The PR-Agent repository, which runs Qodo's hosted review bot on its own pull requests), GitHub issue, published Sep 10, 2026
  29. Qodo No Longer Reviews This Repository, and Two Documents and the Review Script Still Assume It Doesptr727, repository owner (self-described: ProjectTemplate and sibling public repositories run by one maintainer), GitHub issue, published Sep 9, 2026
  30. Comment on: GPT-6 Astra in code review: Gains, privacy, and costoscarestudillom, Hacker News commenter (self-described: Team that added AI review because senior engineers kept repeating the same comments), Hacker News comment, published Sep 9, 2026
  31. A Huge Help for Code Reviews, Bug Catching, and Pull RequestsJavier C., reviewer on G2 (self-described: Reviews of their own code at work, alongside Copilot), G2 review, syndicated on AWS Marketplace, published Sep 3, 2026
  32. PR-Agent Learns from Past Merges and Integrates Seamlessly with GitHubDheeraj J., reviewer on G2 (self-described: Self-hosted PR-Agent on GitHub), G2 review, syndicated on AWS Marketplace, published Sep 1, 2026
  33. A Useful Extra Check Before Merging CodePratap J., reviewer on G2 (self-described: Reviews of their own changes before calling them finished), G2 review, syndicated on AWS Marketplace, published Aug 30, 2026
  34. Speeds Up Coding and Reviews, But Struggles with Complex CodeKarandeep S., reviewer on G2 (self-described: Reviews of their own code), G2 review, syndicated on AWS Marketplace, published Aug 27, 2026
  35. Big-Picture Code Reviews with Qodo, Great for Catching Logic and Dependency IssuesVivaan K., reviewer on G2 (self-described: Data engineer reviewing pipeline changes against team coding standards), G2 review, syndicated on AWS Marketplace, published Aug 25, 2026
  36. Qodo Gives My Code a Reliable Second Pair of EyesSagar K., reviewer on G2 (self-described: Fast-moving personal projects written with coding agents), G2 review, syndicated on AWS Marketplace, published Aug 23, 2026
  37. Invaluable for Code Quality and Repository TrackingZac O., reviewer on G2 (self-described: Reviews across several repositories), G2 review, syndicated on AWS Marketplace, published Aug 23, 2026
  38. Smart Multi-Agent Reviews That Deliver High-Quality FeedbackChanchal R., reviewer on G2 (self-described: Qodo multi-agent review with rules and indexed repository history), G2 review, syndicated on AWS Marketplace, published Aug 8, 2026
  39. Essential Tool for Code Review with Easy SetupAkash S., reviewer on G2 (self-described: Reviews of their own pull requests on larger features), G2 review, syndicated on AWS Marketplace, published Jul 31, 2026
  40. docs(process): qodo is BACK - check it on every PR, alongside codexkstruck, repository owner (self-described: MMPoolsV3 repository), GitHub pull request, published Jul 30, 2026
  41. Finds Bugs Fast, Lets Reviewers Focus on the Bigger PictureOrlando T., reviewer on G2 (self-described: Code review), G2 review, syndicated on AWS Marketplace, published Jul 29, 2026
  42. My Code Reviewer While Building sphragio.ai SoloReinfried B., reviewer on G2 (self-described: Solo founder of sphragio.ai, a Django and FastAPI SaaS), G2 review, syndicated on AWS Marketplace, published Jul 27, 2026
  43. Add Qodo acknowledgement actionSubhramit Basu Bhowmick, maintainer at JabRef (JabRef open source reference manager, a Java desktop app with a large share of AI-assisted outside pull requests), GitHub pull request on JabRef/jabref, published Jul 26, 2026
  44. Rename kanin-loop → review loop (multi-reviewer: CodeRabbit + Qodo)mbjorke, repository owner (self-described: timelog-extract repository with a multi-reviewer loop), GitHub issue comment, published Jul 22, 2026
  45. Fast and Confident, but Accuracy Falls Short on Larger TasksBen C., reviewer on G2, self-described, G2 review, syndicated on AWS Marketplace, published Jul 10, 2026
  46. Underwhelming Experience with Limited Features for the PriceMark H., reviewer on G2 (self-described: Team members joined through GitHub; context features unavailable), G2 review, syndicated on AWS Marketplace, published Jun 13, 2026
  47. Comment on: Open Code Review - An AI-powered code review CLI toollukeasrodgers, Hacker News commenter (self-described: Employer team), Hacker News comment, published Jun 5, 2026
  48. [Enhancement] Enable AI-powered PR code review and suggestions (PR-Agent via opensearch-build)Mohit Kumar, maintainer of cross-cluster-replication at OpenSearch (OpenSearch project repositories maintained by Amazon engineers, sharing reusable review workflows from opensearch-build), GitHub issue on opensearch-project/cross-cluster-replication, published May 7, 2026
  49. Understanding the Limits of Automated Evaluation for Code Review Bots in PracticeVeli Karakaya, Bilkent University, with Baykal Mehmet Uçar of Beko (self-described: Beko's production pull request workflow on Azure DevOps, where developers must label every bot comment before merge), Research paper, arXiv (EASE 2026), published Apr 27, 2026
  50. How I Added AI Code Reviews to Our GitLab Pipeline Using PR-AgentAmbrish Vadnerkar, Medium (self-described: Their team's GitLab project, PR-Agent as an advisory CI job on every merge request with OpenAI, then Gemini 2.5 Flash), Personal blog post, Medium, published Apr 6, 2026
  51. Comparing 9 Code Review Tools - Which is Best?Mike Cann, Convex (self-described: Ten seeded pull requests on a React, Vite, and Convex project, scored against nine review bots; video version published March 10, 2026), Engineering blog post, Convex Stack (Convex engineering blog), published Mar 10, 2026
  52. Comment on: A real-world benchmark for AI code reviewmdeeks, Hacker News commenter (self-described: Comparing review tools on price and quota), Hacker News comment, published Feb 4, 2026
  53. Comment on: A real-world benchmark for AI code reviewmattvv, Hacker News commenter (self-described: Team whose developers open many pull requests a day), Hacker News comment, published Feb 4, 2026
  54. We Ran 11 AI PR Bots (and kept 1)swishfishdemolish, Indie Hackers member (self-described: Team that ran 11 review bots on a production repository and kept Cursor Bugbot), Indie Hackers post, published Jan 29, 2026
  55. Is it possible to remember code suggestions from the previous push?dbudwin, PR-Agent user (self-described: Team running open source PR-Agent in GitHub Actions on every push, with Claude Sonnet 4.5 on Amazon Bedrock), GitHub issue, published Jan 27, 2026
  56. Comment on: Devin Review: AI to Stop Slopjoshgachnang, Hacker News commenter (self-described: CTO and founder of Flourish Health, adolescent mental healthcare), Hacker News comment, published Jan 21, 2026
  57. How We Exploited Qodo: From a PR Comment to RCE and an AWS Admin Key - Leaked TwiceNils Amiet, Senior Security Researcher at Kudelski Security (self-described: Security research on Qodo Merge and Qodo Merge Pro), Security research, Kudelski Security Research Center, published Jan 15, 2026
  58. Comment on: Graphite is joining Cursoroutofdistro, Hacker News commenter (self-described: Employer team), Hacker News comment, published Dec 20, 2025
  59. Comment on: Graphite is joining Cursorarchon810, Hacker News commenter (self-described: Employer repositories, several AI reviewers at once), Hacker News comment, published Dec 19, 2025
  60. Remove Qodo's PR agent CI workflow configlukhnos, OpenVanilla collaborator (self-described: OpenVanilla fcitx5-mcbopomofo input method project), GitHub pull request, published Dec 10, 2025
  61. Incorrect Inline Code Suggestion Formatting in Azure DevOpsspanwair, PR-Agent user (self-described: PR-Agent in an Azure DevOps pipeline), GitHub issue, published Nov 20, 2025
  62. Effortless Communication for Dev TeamsJonah V., reviewer on G2 (self-described: Development team that found code review a time sink), G2 review, syndicated on AWS Marketplace, published Nov 14, 2025
  63. Comment on a Hacker News thread (story no longer available)esafak, Hacker News commenter (self-described: Evaluating tools for own use), Hacker News comment, published Nov 10, 2025
  64. Comment on: Qodo CLI agent scores 71.2% on SWE-bench Verifiedlirantal, Hacker News commenter, self-described, Hacker News comment, published Aug 13, 2025
  65. Comment on: Qodo CLI agent scores 71.2% on SWE-bench Verifiedafro88, Hacker News commenter (self-described: Individual use), Hacker News comment, published Aug 12, 2025
  66. Better Code Reviews with AI? GitHub Copilot and Qodo Merge TestedViktor Farcic, DevOps & AI Toolkit (self-described: Hands-on comparison on one of their own pull requests, with Copilot code review as the existing reviewer), Video, YouTube (DevOps & AI Toolkit), published Jun 30, 2025
  67. Automated Code Review In PracticeUmut Cihan, Bilkent University, with co-authors from Beko (self-described: Beko, a multinational home appliance maker: about 238 practitioners across ten projects had a reviewer built on PR-Agent; three projects and 4,335 pull requests analyzed), Conference talk page, ICSE 2025 Software Engineering in Practice track, published May 2, 2025
  68. Comment on: Effective AI code suggestions: less is moregronky_, Hacker News commenter (self-described: Employer that had recently started using Qodo), Hacker News comment, published Jan 29, 2025

Jan 2025 to Oct 2026.

53 sources from practitioners, by month and sentiment.

The full readExplore all Qodo Signals6 praise themes, 5 concerns, 2 points of agreement and 2 disagreements across all three groups, with all 68 sources open to inspect.

Where It Fits

Where does Qodo belong in the stack?

Open Where It Fits: the Stack Field and the full map

Qodo is Core across Validate, supports Build and Ship, and touches Plan and Operate.

Business area
Product & Engineering
Function
Software Development
Operating stack
Agentic Software Development
Primary role
AI Code Review & Review Standards
Qodo beside the rest of its stackThe Stack Field, every product on every layer
FootprintQodo across the Agentic Software Development stack. Bar height is its role on each layer.

  1. Plan

    1. Work planning: Touches

      Work planning, touches. Qodo's review agent checks a pull request against its linked ticket or spec and reports requirement gaps. Planning stays in the issue tracker.

  2. Build

    1. Codebase context: Supports

      Codebase context, supports. The Context Engine indexes repositories, dependencies, and pull request history. The Software Map and Codebase Wisdom skill give that context to reviewers and coding agents.

    2. Coding: Touches

      Coding, touches. Findings include suggested fixes and a fix command. A coding agent writes the feature.

    3. Agent runtime: Touches

      Agent runtime, touches. The Agentic Toolbox connects Qodo review to Claude Code, Codex, Kiro, and MCP clients. Qodo set October 30, 2026 as the IDE plugin's retirement date.

    4. Tool execution: Supports

      Tool execution, supports. Reviews pull context from ticketing, documentation, and design tools such as Jira and Figma. Qodo exposes tools through MCP and its CLI.

  3. Validate

    1. Debugging: Touches

      Debugging, touches. Pull request chat answers questions about a change. Qodo has no runtime data, so it is not a debugging tool.

    2. Testing: Touches

      Testing, touches. Findings can point to missing tests. Qodo deprecated the v1 test generation command on September 1, 2026 with no replacement, so your test suite stays separate.

    3. Code review: Core

      Code review, core. This is Qodo's primary role. Its agents review each pull request for bugs, rule violations, requirement gaps, and cross-repository breaking changes, and rank findings by relevance learned from past reviews.

    4. Application security: Supports

      Application security, supports. Review agents flag security-relevant bugs and Review Standards can encode security rules. Qodo is not a SAST, secrets, or dependency scanner.

  4. Ship

    1. Source control: Supports

      Source control, supports. PR Triage ranks open pull requests by priority and wait time across repositories. The Git platform remains the system of record.

    2. CI/CD: Touches

      CI/CD, touches. Qodo posts reviews to the pull request and does not build or deploy. Teams that gate merges on it report parsing comments because Qodo publishes no check run.

  5. Operate

    1. Observability: Touches

      Observability, touches. Incident response stays outside Qodo, which has no production telemetry.

Market mapThe layers Qodo covers, and who else owns each one.

  1. Code review & quality

    CoreValidate

  2. Codebase context & intelligence

    SupportsBuild

  3. Tool execution & integrations

    SupportsBuild

  4. Application security

    SupportsValidate

    • Snyk
    • Semgrep

  5. Source control & change management

    SupportsShip

    • GitHub
    • GitLab

7 more layers on the full map. Core, Supports and Touches describe coverage. They say nothing about quality.

More on Qodo

  1. GILD ReviewGILD's evaluation, with the buying facts
  2. Research2 pieces
  3. Guides4 guides
  4. Operator Insights3 operator accounts
  5. Where It FitsOn the Agentic Software Development stack
  6. Related Products6 products