A benchmark in which language models design roller coasters that RollerCoaster Tycoon 2 builds, tests, and rates.
{"dir":0}{"deltas":[{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":0,"piece":"flat","slope_out":0},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":0,"piece":"end_station","slope_out":0},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":0,"piece":"begin_station","slope_out":0},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":0,"piece":"middle_station","slope_out":0},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":16,"piece":"up_25","slope_out":2},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":64,"piece":"up_60","slope_out":4},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":8,"piece":"flat_to_up_25","slope_out":2},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":32,"piece":"up_25_to_up_60","slope_out":4},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":32,"piece":"up_60_to_up_25","slope_out":2},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":8,"piece":"up_25_to_flat","slope_out":0},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":-16,"piece":"down_25","slope_out":6},{"bank_out":0,"dir_out":0,"dx_tiles":-1,"dy_tiles":0,"dz":-64,"piece":"down_60","slope_out":8},{"bank_out":0,"dir_out":0,"dx… (7397 chars)
{"dir":1}{"deltas":[{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":0,"piece":"flat","slope_out":0},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":0,"piece":"end_station","slope_out":0},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":0,"piece":"begin_station","slope_out":0},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":0,"piece":"middle_station","slope_out":0},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":16,"piece":"up_25","slope_out":2},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":64,"piece":"up_60","slope_out":4},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":8,"piece":"flat_to_up_25","slope_out":2},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":32,"piece":"up_25_to_up_60","slope_out":4},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":32,"piece":"up_60_to_up_25","slope_out":2},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":8,"piece":"up_25_to_flat","slope_out":0},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":-16,"piece":"down_25","slope_out":6},{"bank_out":0,"dir_out":1,"dx_tiles":0,"dy_tiles":1,"dz":-64,"piece":"down_60","slope_out":8},{"bank_out":0,"dir_out":1,"dx_tiles":0,"d… (7346 chars)
← openrouter_deepseek_deepseek-v4-flash-0731 · 20260731-deepseek-v4-flash-design · all runs