FrontierChallenge: Evaluating Scientific Workflow Completion

FrontierChallenge releases 97 representative tasks drawn from a 300-workflow cross-domain benchmark spanning quantum chemistry, molecular dynamics, materials characterization, analytical chemistry, life science, and electrochemistry. The best-performing configuration completes only 20 of 97 tasks (20.6% pass rate), and 75.5% of failing Claude Code trajectories still verbally claim completion.