Benchmarks, evaluations, and methods from Composite, the team building AI for professional computer work.https://research.composite.comhttps://research.composite.com/composite-benchhttps://research.composite.com/composite-benchThe first long-horizon, browser-based computer-use benchmark with certified optimality and verified compute. In our evaluation, GLM-5.2 lands 11 points behind a two-way Claude tie and beats every closed model we measured.Fri, 24 Jul 2026 00:00:00 GMT