Research
Rogue Constellation scores 71.21% on Terminal-Bench 4.0
Rogue's flagship coding agent posted the top published score on Terminal-Bench 4.0, ahead of Claude Opus 5.5 and GPT-6 Astra.
Rogue launched Rogue Constellation, its flagship coding agent for complex software work. It orchestrates the best frontier models.
On Terminal-Bench 4.0 it scored 71.21%, against 57.9% for GPT-6 Astra, 57.4% for Gemini 4 Argon and 55.8% for Fable 5.1. On Terminal-Bench-Science 0.1, Constellation scored 71.43%, against 64.6% for GPT-6 Astra. The announcement shows how it chooses, what it costs and three games it built.
