Rogue
Constellation
Our flagship coding agent.
3 min read
Today, we are introducing Rogue Constellation, a new multi-agent system for complex software work. It orchestrates the best frontier models, delivering the world's most powerful agentic coding experience.
An implementation can look convincing and still miss a requirement, fail an edge case, or solve the wrong problem. Exploring alternatives and deciding which result to trust takes substantial effort.
Constellation starts five agents from the same state of your project. Each writes code and runs tests independently. It then compares their approaches and the evidence produced along the way to construct a single solution.
We evaluated it on two demanding benchmarks. Here’s how it performed.
Terminal-Bench 4.0
Rogue Constellation 71.21%
GPT-6 Astra 57.9%
Gemini 4 Argon 57.4%
Claude Fable 5.1 55.8%
Terminal-Bench Science 0.1
Rogue Constellation 71.43%
GPT-6 Astra 64.6%
Gemini 4 Argon 57.6%
Claude Fable 5.1 52.6%
See it in action
Play three games built by Constellation, with graphics and sound generated entirely in code.
Efficiency
Cost was one of the main concerns with Rogue DeepThink. We took that feedback seriously and made efficiency a central design priority for Constellation.
Usage reflects the combined work of all agents, including attempts that are discarded. The chart below compares accuracy and spending at different effort levels.
Terminal-Bench 4.0
- Constellation
- DeepThink Preview
- Fable 5.1
- GPT-6 Astra
Availability and pricing
Constellation is available today on every paid Rogue plan.
Rogue ConstellationUSD per million tokens
- Input
- $10
- Output
- $25
- Cached input
- $2
These rates determine how much of your plan’s allowance a task consumes. Usage counts the tokens of all agents.