RUN 20260725-codex-scored (design + open note)

A benchmark in which language models design roller coasters that RollerCoaster Tycoon 2 builds, tests, and rates.

Standings

design mode — models build from scratch

modelscoreintensitynausea similaritybesttokens / cost
1st openrouter_openai_gpt-5.1 no successful coaster $1.56

score = excitement, scaled down when similarity to a stock library design exceeds 0.5 (an exact or mirrored copy scores 0)

openrouter_openai_gpt-5.1full detail →

Round 1 not rated trace (271) →

built, but the ride was never rated

track program (75 pieces)
{
  "pieces": [
    "begin_station",
    "middle_station",
    "middle_station",
    "middle_station",
    "end_station",
    {
      "chain": true,
      "t": "flat_to_up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25_to_flat"
    },
    "flat",
    "flat",
    "flat_to_down_60",
    "down_60",
    "down_60_to_flat",
    "flat_to_right_bank",
    "banked_right_turn_5",
    "right_bank_to_flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat_to_up_25",
    "up_25",
    "left_vertical_loop",
    "down_25_to_flat",
    "flat",
    "right_corkscrew_up",
    "right_corkscrew_down",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "left_turn_3",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat_to_left_bank",
    "left_helix_down_small",
    "left_bank_to_flat",
    "s_bend_left",
    "s_bend_left",
    "right_turn_3",
    "right_turn_3",
    "s_bend_right",
    "left_turn_3",
    "right_turn_3",
    "s_bend_right",
    "right_turn_3",
    "left_turn_3",
    "right_turn_3",
    "right_turn_3",
    "flat"
  ],
  "ride_type": 51,
  "start": {
    "dir": 1,
    "x": 50,
    "y": 60
  }
}
park screenshot, openrouter_openai_gpt-5.1 round 1

← all runs