Recent findings from the ORCA Benchmark reveal that leading AI models fail nearly 40% of simple math tasks, raising concerns about their reliability.

Apodex's TRACES benchmark redefines AI evaluation by allowing systems to tackle real-world scientific challenges and adapt to dynamic environments.

Insilico Medicine unveils a pioneering Drug Discovery Benchmark as a Service, allowing organizations to assess the performance of AI models in real-world drug discovery applications.

Druid AI has released its 2026 AI Adoption Benchmark Report, revealing intriguing insights about AI's actual performance in various industries. This report highlights the disparity between AI's perceived potential and its practical use in businesses.

Druid AI's latest 2026 AI Adoption Benchmark Report reveals the significant divergences between AI expectations and actual enterprise usage across various sectors.

ORCFLO has introduced the ORCFLO Index, a new benchmark designed to evaluate AI models for crucial business tasks, offering insights into cost, time, and quality.

Discover APIEval-20, KushoAI's innovative open benchmark for AI-driven API test generation, designed to enhance performance evaluation for developers.

Sup AI has set a new record by scoring 52.15% on the challenging Humanity's Last Exam, outpacing leading AI models and showcasing its innovative orchestration system.