Claude Sonnet 5.5 is now available in Devin

Cognition1 min read

On FrontierCode 1.1, Sonnet 5.5 is a significant improvement from Sonnet 5 at 64.4%, surpassing even Fable 5.1 at extra high reasoning effort.

Anthropic reports more than 30% faster output generation and up to 30% lower cost per task than Sonnet 5, with unchanged token pricing.

FrontierCode LeaderboardBenchmarks for how well models meet the standards of high-quality production codebasesView Now

FrontierCode 1.1 ExtendedScore

020406080100Score (%)56.2Claude Sonnet 556.3Gemini 3.7 Flash60.7GPT-6 Sol61.3Grok 4.662.5SWE-263.6Claude Opus 563.6Claude Fable 5.164.4Claude Sonnet 5.564.5GPT-6 Astra64.9Claude Fable 565.3Claude Opus 5.5

Score is a weighted aggregate of rubric items. Solutions that don’t pass blocking criteria receive 0.

FrontierCode 1.1 ExtendedScorevsCost ($)

$0.25$0.50$1$2$5$10$20avg cost (USD) per rollout4550556065score (%)Opus 5.5Sonnet 5.5GPT-6 AstraFable 5.1SWE-2GPT-6 SolSonnet 5

Score: a weighted aggregate of the rubric items. Solutions that don’t pass blocking criteria receive 0.

Cost ($): the mean USD spend per rollout. Each dot is a reasoning effort.

Download Devin Desktop · Install Devin CLI