OpenAI recently released its newest model, GPT 5.5.

At Rogo, we operate a model-agnostic platform, continuously benchmarking and integrating leading models from OpenAI, Google, and Anthropic. As part of this effort, we’ve begun incorporating GPT 5.5 into our agent harness, Felix, to evaluate performance across real-world financial workflows.

At the core of this evaluation is Rogo’s Big Finance Benchmark, an internally curated set of real financial tasks designed by our ex-finance team. Each task mirrors the work that lands on a professional’s desk: valuing companies, benchmarking peers on specific metrics, or building theses across disparate documents. Because these problems come from real workflows, not synthetic prompts, Big Finance captures what truly differentiates top models in this domain.

In initial testing, GPT 5.5 shows improved financial reasoning over prior models, with stronger multi-step analysis, more consistent handling of complex logic, and outputs that better reflect how financial professionals think.

We’re excited to bring GPT 5.5 to the 35,000 finance professionals who use Rogo today.