Kodus
Open-source self-hostable AI code review tool, officially claiming that you can use your own model keys with no markup, unlimited PRs, the ability to identify issues across files and enforce team-specific custom rules.
Updated: 2026-10-26
What is Kodus?
Kodus is an open-source AI code review tool, and its official website directly positions itself as an "open-source alternative to CodeRabbit". Its model is: it can be self-hosted for free, or you can use the official cloud version with your own model keys, paying the original price of the model manufacturer without any price increase by the platform. The official claims that it has been used by more than 5,000 teams in 66 countries. The cloud version offers a 14-day free trial, including up to 35 PR reviews, and no credit card is required.
Who is it for?
- Engineering teams with a high volume of PRs that manual review can't keep up with, and who want every submission to be pre-checked by a machine.
- Enterprises and compliance teams that have requirements to keep code within the intranet and must self-host the code review service.
- Leaders who want to control model costs and do not want the code review tool to mark up the price of model fees.
- Engineering efficiency teams that have clear coding standards and want machines to consistently enforce them.
Key features
6Open source core
The core is open-sourced under the AGPL license, and the official states that the code can be read and run on your own infrastructure.
Any model, zero markup
Connect to any model with your own API key, pay the vendor's original price, and the platform emphasizes zero markup.
No PR rate limits
The official states that every push will be reviewed, and all plans have no limit on the number of PRs.
Catches bugs across files and repos
Detects issues across files and repositories, rather than just looking at a single diff.
Enforces your team's rules
Checks against your team's custom rules, and the official also mentions that business rules can be verified.
Talk to Kody in the PR
You can directly chat with the review bot Kody in the PR, and technical debt tracking is supported.
How to use Kodus
- For self-hosting, the official states that it can be deployed using Docker Compose or Helm.
- First, test it on one or two repositories, adjust the rules, and then roll it out fully.
- In the self-hosted key mode, the model cost is entirely determined by your usage.
- Estimate the number of PRs and the context length first before scaling up.
- The official website has a dedicated section titled "You'll hate us if...".
- It is recommended to check it out first to determine in advance if it is suitable for your team.
- Review suggestions are for reference only.
- changes related to security and core links still require manual confirmation.
- licensing and commercial terms are subject to the official website, and the rules may be adjusted.
- Please verify them again before official use.
Traffic & popularity
Data month Sep 2026
Kodus receives about 13.1K visits per month and ranks #2,464,833 worldwide.
Monthly visits
12 monthsTraffic sources
- Direct41.1%
- Search33.8%
- Referrals10.5%
- Social6.5%
- Mail2.7%
- Other5.5%
Top countries
- Brazil18.8%
- Portugal16.1%
- United States9.2%
- India6.4%
- Vietnam4.7%
Estimated from third-party traffic data.
Similar tools
TabbyOpen-source self-hosted AI coding assistant offering code completion, a question-answering engine, and in-line chat. The official states it supports deployment on consumer-grade GPUs, requires no external databases or cloud services, and keeps data within your own environment.
Hermes AgentNous Research's open-source and self-hostable AI agent, officially available for free under the MIT license, supports persistent memory, automatic skill generation, multi-platform access, and isolated sandboxes. It can be downloaded for desktop or terminal installation.
AgenticReviewAgenticReview evaluates how effectively teams guide AI agents, extending assessment beyond code quality, security, and testing. According to the company, its benchmark spans more than 30 million commits and approximately 100,000 repositories, reflecting real-world development at scale.
Augment CodeFocusing on an organization-scale intelligent agent software development platform, the official core product is a continuous delivery system that links the entire process from requirements to launch, and is equipped with automated links such as code review, vulnerability repair, and incident response.
KiloOpen-source AI coding agent, covering VS Code, JetBrains, CLI and cloud, supporting access to more than 500 models and zero markup for self-provided keys, and using local models can protect code privacy.
WeWebWeWeb is an IT-approved AI application builder. Officially, it claims to be built with AI and edited with no-code, can be one-click published to a CDN or exported for self-hosting, emphasizes controllability and scalability, and is suitable for teams that need compliance.