AIOpus 5.5 reportedly overtakes Fable 5 on FrontierCode 1.1
Cognition reports a 65.3% score on Extended for Opus 5.5, claiming the model takes the lead at a fraction of Fable 5’s cost.
TLDR
Cognition says Opus 5.5 scored 65.3% on Extended in FrontierCode 1.1, its benchmark for real-world engineering tasks that grades mergeability and quality. The company says that displaces Fable 5 from the leading spot at a fraction of the cost. It also invites users to try Opus 5.5 in Devin Desktop and Devin CLI.
Combined views
2.6K
1 Source, first seen 11h ago
Opus 5.5 reportedly overtakes Fable 5 on FrontierCode 1.1
Cognition reports a 65.3% score on Extended for Opus 5.5, claiming the model takes the lead at a fraction of Fable 5’s cost.
TLDR
Cognition says Opus 5.5 scored 65.3% on Extended in FrontierCode 1.1, its benchmark for real-world engineering tasks that grades mergeability and quality. The company says that displaces Fable 5 from the leading spot at a fraction of the cost. It also invites users to try Opus 5.5 in Devin Desktop and Devin CLI.