A new platform called Optima has been launched, allowing users to create custom AI benchmarks using their own data and workflows. This approach enables more accurate comparisons between AI models, taking into account factors like cost and processing time in addition to raw performance. The platform is particularly useful for agent-based applications, where these metrics can be more relevant than raw pricing. Optima's approach aims to address a key limitation in traditional AI benchmarking methods. This development matters because it could lead to more effective and efficient AI model selection for real-world applications.