OpenAI's newest GPT-5.6 models, Sol, Terra, and Luna, are now available in Rogo. We evaluated each model across real financial workflows to understand where it performs best.

How Sol, Terra, and Luna performed

Big Finance Benchmark - Final Answer Accuracy 

A few takeaways:

• Sol delivers the strongest overall performance, achieving the highest final-answer accuracy in the evaluation at 53%.

• Terra offers the best balance of quality, speed, and cost, matching GPT-5.5 on rubric score while improving final-answer accuracy.

• Luna is the lightweight option in the GPT-5.6 family, using the fewest output tokens of the three models, making it well suited for high-volume, cost-sensitive workloads.

As always, no single model is best across every workflow. The right choice depends on the task, balancing reasoning quality, speed, and cost. That's why Rogo continuously benchmarks new models before making them available.