{"name":"codai benchmarks — romanian","scope":"romanian","license":"https://creativecommons.org/licenses/by/4.0/","attribution":"CC BY 4.0, attribution: codai (https://codai.ro/benchmarks)","source":"https://codai.ro/benchmarks/romanian","selection":"One run per benchmark and model: the completed run with the largest sample; ties go to the most recent finish.","units":{"successRate":"fraction 0..1","costUsd":"USD","costPerSuccessUsd":"USD"},"lastRefresh":"2026-06-27T00:38:28.794Z","benchmarks":[{"id":"cada1ba6-f570-41cb-b662-1792efbe87f1","slug":"romath","name":"RoMath","description":"Romanian high-school mathematics benchmark from RoMathExam (Cuclea et al. 2026, 10,592 problems from official Bacalaureat exams 1895-2025, MIT license). Stratified sample across M1/M2/M3/M4 tracks. Scored by an LLM judge (sonnet-4-6) comparing the candidate's final answer to the official `barem` — math can't be exact-matched. Codai-unique angle: no major LLM router publishes Romanian-specific scores.","category":"romanian","taskCount":100,"referenceUrl":"https://arxiv.org/abs/2409.11074","resultKind":null,"runs":[{"id":"69df94fe-4121-45fe-a35f-5169ae6bd607","modelTested":"claude-opus-4-7","successRate":"0.9900","costUsd":"1.6449","costPerSuccessUsd":"0.0166","completedTasks":100,"totalTasks":100,"finishedAt":"2026-05-27T09:52:36.462Z"},{"id":"8e15d8cd-b554-41ba-9909-0602dfa634ec","modelTested":"codai","successRate":"0.9800","costUsd":"3.7393","costPerSuccessUsd":"0.0382","completedTasks":100,"totalTasks":100,"finishedAt":"2026-06-27T00:38:28.794Z"},{"id":"56624864-8b68-4dad-ac44-a1c8c065aee8","modelTested":"claude-opus-4-8","successRate":"0.9800","costUsd":"2.3230","costPerSuccessUsd":"0.0237","completedTasks":100,"totalTasks":100,"finishedAt":"2026-06-27T00:28:21.954Z"},{"id":"582d7abe-3409-4b90-afbe-4d7043b30fe2","modelTested":"claude-haiku-4-5","successRate":"0.9600","costUsd":"0.9862","costPerSuccessUsd":"0.0103","completedTasks":100,"totalTasks":100,"finishedAt":"2026-06-15T20:23:53.938Z"},{"id":"8fa5af57-a670-4554-bca4-b430c1855a8a","modelTested":"claude-sonnet-4-6","successRate":"0.9600","costUsd":"1.4620","costPerSuccessUsd":"0.0152","completedTasks":100,"totalTasks":100,"finishedAt":"2026-05-27T09:50:26.267Z"},{"id":"661384ea-f29b-4c57-a8df-257505857651","modelTested":"gpt-5-mini","successRate":"0.8000","costUsd":"3.3954","costPerSuccessUsd":"0.0424","completedTasks":100,"totalTasks":100,"finishedAt":"2026-05-27T09:57:14.267Z"},{"id":"4511f7e7-1a71-4229-8ec1-6e04c948f101","modelTested":"codai-alpha-1-0","successRate":"0.1200","costUsd":"1.7195","costPerSuccessUsd":"0.1433","completedTasks":100,"totalTasks":100,"finishedAt":"2026-06-15T22:02:23.722Z"},{"id":"0fb59bd8-56c1-4a13-99e4-4a40347adaf4","modelTested":"claude-fable-5","successRate":"0.0000","costUsd":"0.0000","costPerSuccessUsd":null,"completedTasks":100,"totalTasks":100,"finishedAt":"2026-06-15T20:13:47.859Z"}]}]}