loading
loading
OpenAI released GPT-6.1 Sol. It is live now in Devin Desktop and Devin CLI. On FrontierCode 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.
document fetched 2026-10-24 10:06