Skip to main content

Claude Opus 5.5 is now available in Devin

By Steven Van ·

Claude Opus 5.5 tops Devin's FrontierCode 1.1 benchmark at under a tenth of Fable 5's cost per task, in Devin Desktop and CLI.

Devin now offers Claude Opus 5.5 in Devin Desktop and Devin CLI, according to its announcement.

Benchmark result

Opus 5.5 ranks first on FrontierCode 1.1, Devin's proprietary benchmark that grades models on real-world engineering tasks for quality and mergeability. It edges out Fable 5 and GPT-6 Astra. Per task, it costs less than a tenth of Fable 5 and about a sixth of Astra.

Where it performs well

In Devin's evaluations, Opus 5.5 is particularly strong on tasks where the right answer is to follow existing repo conventions. It reaches a fix with far less exploration than Fable 5, roughly 4x faster. Devin also describes it as disciplined about verification and about keeping diffs scoped to the task.

The FrontierCode leaderboard also lists SWE-2, Fable 5.1, Grok 4.6, GPT-5.6 Sol, Opus 5 and Kimi K3.

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

Original source

Read Claude Opus 5.5 is now available in Devin on Creators Toolbox