Multi-functional LLMs
Claims, underwriting, actuarial, coverage and document tasks. Compare the whole system, including tools and harnesses.
Insurance benchmarks.
Choose the work you want to compare. Each benchmark keeps its own questions, scoring, cost basis and limitations.
Decision Models
Image Inputs, General Text and Yes/No views. Filter by open weights, local runtime and measured deployment; choose models and compare accuracy, cost, completion time and 1–100 question scaling.
Read the latest findings →Original decision-model comparison
The earlier submission-evidence report remains available. Its core and exploratory measurements are also selectable in the updated Decision Models page.
Development evidence and independently reviewed evaluations are labelled separately. Private prompts and answer keys are not published.