GPT-6 Sol and GPT-6 Luna are now available in Devin
By Steven Van ·
Both OpenAI models are live in Devin Desktop and CLI, with Luna coming in under $0.10 per task on Devin's FrontierCode 1.1 benchmark.
OpenAI's GPT-6 Sol and GPT-6 Luna are now live in Devin Desktop and Devin CLI, according to Devin's announcement.
Benchmark results
Devin's numbers come from FrontierCode 1.1, its proprietary benchmark that grades models on real-world engineering tasks based on quality and mergeability.
- GPT-6 Sol matches GPT-5.6 Sol's score at 61% lower cost per task. It lands within four points of GPT-6 Astra at less than half its cost.
- GPT-6 Luna scores above GPT-5.6 Luna at about a quarter of the cost. At under $0.10 per task, it is the cheapest model on the leaderboard.
How the models work
Devin says the gains show up in behaviour as well as cost. In its evaluations, GPT-6 Sol reads about 17% less context than GPT-5.6 Sol to reach the same score, and rarely leaves a repo with failing tests after its last edit.
GPT-6 Luna's biggest gains come at low and medium reasoning effort, where it writes real regression tests instead of mock-based ones.
The models are listed on Devin's FrontierCode leaderboard alongside Opus 5.5, GPT-6 Astra, SWE-2 and the GPT-5.6 models.
