Greptile 6.3/10 Fair · T3
bestaiq
// CODE

Greptile Review (2026): The AI Code Reviewer With a Contested Benchmark Claim

6.3/10
Fair · T3

A repository-aware code reviewer with real enterprise features, but its headline accuracy claim is self-published and worth checking against your own codebase rather than taking at face value.

SUB-SCORE SPINE
7.0
CAP
5.0
VAL
5.5
EAS
7.5
PRI
5.5
SUP
7.0
ECO
Independent · ad-free verdicts · we may earn affiliate commissions — this never affects our scores.
FIG · QUICK ANSWER

Greptile is an AI agent that reviews pull requests on GitHub and GitLab. Instead of only looking at the lines that changed, it builds a graph index of the whole repository so it can catch issues that span multiple files. It launched in 2023, went through Y Combinator's Winter 2024 batch, and has since raised a $25 million Series A led by Benchmark Capital.

TABLE · AT A GLANCE
BestAIQ score 6.3 / 10 (Fair) [liveFacts] ✓Jul'2026
Category AI code review (developer tooling) [liveFacts] ✓Jul'2026
Starting price Free tier available; Pro from $30/seat/month [liveFacts] ✓Jul'2026
Founded 2023 [liveFacts] ✓Jul'2026
Headquarters San Francisco, California [liveFacts] ✓Jul'2026
Self-hosting Available for Enterprise (Docker Compose or Kubernetes) [liveFacts] ✓Jul'2026
Git hosts GitHub Cloud/Enterprise, GitLab Cloud/Self-Managed [liveFacts] ✓Jul'2026

//Strengths & things to watch

STRENGTHS
  • Indexes the entire repository as a graph, so it can flag issues that span multiple files instead of just the lines in a diff.
  • Offers a free tier for individual developers and free access for qualifying MIT or Apache licensed open source projects.
  • Supports self-hosted, air-gapped deployment with a choice of LLM provider for teams with strict data requirements.
  • Routes flagged issues directly to coding agents like Claude Code, Cursor, Codex, or Devin for one-click fixes.
  • States SOC 2 Type II compliance and lets customers opt out of AI training on their code from account settings.
THINGS TO WATCH
  • Its 82% bug catch rate claim comes from a benchmark Greptile designed and scored itself. Independent re-scoring of the same test data by DeepSource and by competitor Augment Code put the figure closer to 45%.
  • Greptile has no meaningful history on major review platforms. G2, Slashdot, and SourceForge each show zero reviews, so there is little independent, aggregated feedback to check against.
  • Pricing changed in March 2026 from a flat unlimited-review plan to 50 reviews per seat with a $1 charge per extra review. Teams reviewing many pull requests a month, especially with agentic workflows, may pay substantially more than before.
  • There is no dedicated IDE extension. Fixes are routed to external coding agents rather than handled inside an editor plugin.
  • Bitbucket and Azure DevOps are not supported.

//How it works

  1. 01
    Connect a repository

    Greptile connects to GitHub Cloud, GitHub Enterprise Server, GitLab Cloud, or GitLab Self-Managed, or runs self-hosted for enterprise customers.

  2. 02
    Index the codebase

    It builds a graph index of files, functions, and dependencies across the repository rather than looking only at the changed lines.

  3. 03
    Review pull requests

    When a PR opens, Greptile posts inline comments, a PR summary, and sequence diagrams, typically within a few minutes, with each comment carrying a confidence score.

  4. 04
    Generate tests (optional)

    Its TREX feature, in public beta, can generate and run unit tests for the changed code in a sandbox, at an added credit cost.

  5. 05
    Route fixes to an agent

    A 'Fix with your Agent' button on each comment sends the file path, line number, and suggested code to Claude Code, Codex, Cursor, Devin, or Conductor.

FIG · SCORE BREAKDOWN
CapabilityCAP
0.30WEIGHT
7.0
ValueVAL
0.20WEIGHT
5.0
Ease of useEAS
0.15WEIGHT
5.5
PrivacyPRI
0.15WEIGHT
7.5
SupportSUP
0.10WEIGHT
5.5
EcosystemECO
0.10WEIGHT
7.0
SCORING PIPELINE — SHOW THE WORK
SOURCES
61
SUB-SCORES
6 DIMS
WEIGHTED
Σ=1.0
EDITORIAL
NONE
VERDICT
6.3/10

//Who it's for

CHOOSE GREPTILE IF…
  • Teams that want repository-wide context in code review, not just diff-level linting.
  • Open source maintainers with MIT or Apache licensed projects who qualify for free usage.
  • Enterprises that need self-hosted or air-gapped deployment with a choice of LLM provider.
LOOK ELSEWHERE IF…
  • Teams with a high volume of pull requests each month, where the per-review overage pricing could get expensive.
  • Buyers who want to check independent review-platform ratings first. Greptile has almost none.
  • Teams that need native Bitbucket or Azure DevOps support today.
TABLE · HOW IT COMPARES
ToolScoreTierFromFreeLink
GreptileTHIS TOOL 6.3 Fair · T3 $30/mo [liveFacts] Yes
Playwright 8.4 EXCELLENT · T1 $0/mo [liveFacts] Yes Review →
Claude Code 7.9 GREAT · T2 $17/mo [liveFacts] No Review →
GitHub Copilot 7.8 GREAT · T2 $0/mo [liveFacts] Yes Review →

//What users say

AI synthesis of external reviews · not on bestaiq

G2
MIXED1 review
[S]
Slashdot
MIXED1 source
[S]
SourceForge
0.0/51 source
[S]
Hacker News
MIXED7 sources
[S]
◆ AI SUMMARY

Synthesized from 5 external reviews. Independent signal (Trustpilot / Reddit / verified aggregators) weighted higher than commission-carrying review sites.

MOST PRAISED
  • Catches real issues before human review reaches the PR, per multiple Hacker News accounts
  • Full-repository context valued over diff-only linting
  • Founders and team members directly engage with technical criticism in public forums
MOST CRITICIZED
  • Self-published benchmark numbers do not match independent re-scoring of the same test data
  • 2026 per-review pricing seen as costly for high-PR-volume or agentic workflows
  • Reports of inconsistent output quality and factual errors in review comments
  • No in-app cancellation button reported by one independent source
TAKEAWAYSentiment lives almost entirely on Hacker News and X rather than aggregated review platforms, so there is no large-sample rating to point to

//Frequently asked

Q1

How much does Greptile cost?

Greptile has a free tier for one developer with 50 review credits a month. The Pro tier is $30 per seat per month with 50 credits per seat and $1 for each additional credit. Enterprise pricing is custom and requires contacting sales.

Q2

Is Greptile's 82% bug catch rate claim independently verified?

No. It comes from a benchmark that Greptile designed, ran, and scored itself in July 2025. When DeepSource and competitor Augment Code re-scored the same five repositories with different judging rules, they reported a catch rate closer to 45% for Greptile, so treat the 82% figure as a vendor claim rather than an independently confirmed result.

Q3

Does Greptile train its AI on customer code?

Greptile's terms allow it to use de-identified, aggregated customer data to train its models, but customers can opt out from their account settings. For self-hosted deployments, no customer data is used for machine learning unless the customer explicitly configures it.

Q4

Can Greptile be self-hosted?

Yes. Enterprise customers can deploy Greptile on-prem using Docker Compose or Kubernetes, with a choice of LLM provider including OpenAI, Anthropic, AWS Bedrock, Azure OpenAI, or GCP Vertex AI. Self-hosting requires a separate license.

Q5

Which git hosts does Greptile support?

GitHub Cloud, GitHub Enterprise Server, GitLab Cloud, and GitLab Self-Managed. It does not currently support Bitbucket or Azure DevOps.

Q6

Is Greptile free for open source projects?

Greptile states it is free for qualifying non-commercial projects licensed under MIT or Apache. Some maintainers have reported being billed anyway, so it is worth confirming terms directly with Greptile before relying on the free tier for a production open source project.

BOTTOM LINE
Greptile

Greptile reviews entire pull requests against a graph of the whole repository, which lets it catch issues a diff-only tool would miss, and it offers real enterprise features like self-hosting and a choice of LLM provider. Its central performance claim, an 82% bug catch rate against competitors, is self-published and was scored much lower when independent parties re-ran the same test, so buyers should verify results against their own codebase rather than the vendor's number. The March 2026 move to per-review pricing also makes it worth modeling actual monthly PR volume before committing to Pro.

Fair · T3 6.3/10

//Related tools

//Featured in

61 sources· Last verified Jul 2026 ✓ VERIFIED