# Grok 4.7 is now available in Devin

By Steven Van · 2026-09-21

Devin's benchmark finds the model strongest on hard Java, Go and Ruby backend tasks, but it scores below Grok 4.6 overall because it over-scopes.

Grok 4.7 is now live in [Devin](<https://creatorstoolbox.com/tools/devin>) Desktop and the Devin CLI, [according to Devin's announcement](<https://devin.ai/blog/grok-4-7>).

## Where it does well

Devin rated the model on FrontierCode 1.1, its proprietary benchmark that grades models on real-world engineering tasks based on quality and mergeability. Grok 4.7 is strongest on hard, multi-module backend tasks in Java, Go, and Ruby. On those tasks it investigates deeply, reads dependency source, and verifies its work end to end with real integration tests.

## Where it falls short

Grok 4.7 scores below Grok 4.6 on the aggregate. Devin attributes this to over-scoping: the model tends to produce larger diffs that include changes the evaluation task did not request.

[![](<https://devin.ai/images/blog/grok-4-7/cover.png>)](<https://devin.ai/blog/grok-4-7>)

![Devin](<https://devin.ai/favicon.svg>)

Devin

AI software engineering agent for building features, fixing bugs, reviewing code, and handling work across repositories.

[View Devin →](<https://creatorstoolbox.com/tools/devin>)

Sources

[Original source](<https://devin.ai/blog/grok-4-7>)

[Read Grok 4.7 is now available in Devin on Creators Toolbox](<https://creatorstoolbox.com/blog/devin-grok-4-7-is-now-available-in-devin>)

---
Canonical source: https://creatorstoolbox.com/blog/devin-grok-4-7-is-now-available-in-devin