An AI agent that builds a graph index of your codebase to review pull requests with full context, available as a cloud service or self-hosted through Docker or Kubernetes.
Our description, from the vendor's documentation, as of 2026-10-14; reasoned from those pages, not observed running.
Models runWhere the vendor's own pages say the product's models run, not a limit: your machine, a container environment you run, the model company's cloud (servers of the company that makes or serves the model), or a cloud provider's model service. For a cloud, read the data and compliance lines of the company running it, linked by its name.
Not stated
Data and compliance
Compliance: Not checked yetWe have not read a page of the right kind for this question yet.: no page that could answer it is recorded as read.
Data: Training use, zero-retention option and data residency not checked yetWe have not read a page of the right kind for this question yet.; default retention not stated on the pages readWe read a page that could answer and found no sentence that does. It says nothing about what the maker does..
Model and where it runs: Hosted by the maker; model provider and where it can run not checked yetWe have not read a page of the right kind for this question yet..
Details: What the maker's own pages state, not a compliance determination, unverified after 2027-01-12: trust page, www.greptile.com.
What the maker's own pages said on the dates shown, not a compliance determination.
Greptile home page(the product's own site, from our list of products; not read for these lines)
The readings behind this page
Our summary of the vendor's pages as of 2026-10-25; confirm on the vendor's page.
Models from different providers in one session
Greptile's blog post on building a code review agent says the company uses both frontier and open source models, selecting the right-sized model for each subtask in code review and testing workflows. (www.greptile.com page, read 2026-10-25)
Several endpoints, built in (no proxy)
Greptile's guide to AI code review for GitLab and GitHub Enterprise servers a team hosts itself says Greptile, in that setup, uses the models the customer picks: AWS Bedrock, Anthropic, Azure OpenAI, OpenAI, or any endpoint compatible with OpenAI. (www.greptile.com page, read 2026-10-25)
Plan, policy and billing
Greptile's pricing page lists self-hosting in the customer's own infrastructure among what the Enterprise plan adds to Pro. (www.greptile.com page, read 2026-10-25)
Greptile's enterprise page says enterprise customers whose data privacy or compliance requirements are strict can have Greptile installed in their own infrastructure, by contacting Greptile's team. (www.greptile.com page, read 2026-10-25)
Greptile's enterprise page says teams can define review configurations from the organization level down to individual developers, tailoring code review to each team's standards. (www.greptile.com page, read 2026-10-25)
Greptile's review tiers page says an Auto rule lets the tool pick the tier for matching pull requests, even when the organization has pinned a tier. (www.greptile.com page, read 2026-10-25)
Greptile's pricing page says the Pro plan has no seat fees and charges $1 per credit, with credits shared across the whole organization. (www.greptile.com page, read 2026-10-25)
Greptile's pricing page says the Starter plan costs nothing for a single developer and comes with 50 monthly credits. (www.greptile.com page, read 2026-10-25)
Greptile's billing page says the company bills using prepaid credits. (www.greptile.com page, read 2026-10-25)
Greptile's billing page says that with Auto, the customer pays for the review tier that ran. (www.greptile.com page, read 2026-10-25)
About the product
Greptile's manual setup page for running Greptile with Docker Compose lists the server's supported operating systems as Ubuntu 20.04 or later, Amazon Linux 2023, Debian 11 or later, and RHEL 8 or later. (www.greptile.com page, read 2026-10-25)
Our reading of vendors' pages, as of the dates shown. Confirm on the vendor's page before relying on any line that cites it. To report an error write to report@crinaro.ai