[](https://devin.ai/blog)

# GPT-6.1 Sol is now available in Devin

CognitionSeptember 29, 20261 min read

OpenAI released GPT-6.1 Sol. It is live now in Devin Desktop and Devin CLI.

On [FrontierCode 1.1](https://cognition.ai/blog/frontier-code-1.1), our proprietary benchmark that grades models on real-world engineering tasks based on quality and mergeability, GPT-6.1 Sol scores 60.4%, in line with GPT-6 Sol (60.7%) and GPT-5.6 Sol (60.6%), but it gets there for a fraction of the cost. At medium reasoning effort it costs $0.31 per task, 81% less than the $1.66 GPT-6 Sol spends at max effort to reach its best score. Across every effort level, GPT-6.1 Sol runs 44% to 57% cheaper per task than GPT-6 Sol while landing within a point of it or ahead.

The bigger change is at the low end. At low effort, GPT-6.1 Sol scores 58.1% for $0.21 per task, up from 50.5% for GPT-6 Sol at the same setting. That is the highest score of any model under $0.30 per task on the leaderboard.

[FrontierCode LeaderboardBenchmarks for how well models meet the standards of high-quality production codebasesView Now](https://cognition.ai/frontiercode)

#### FrontierCode 1.1 ExtendedScore

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

#### FrontierCode 1.1 ExtendedScorevsCost ($)

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](https://devin.ai/download) · [Install Devin CLI](https://docs.devin.ai/cli)