CLAUDE-OPUS-5 · 20260725-opus5-python-probe

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

claude-opus-5 · 1st of 1 compare all models →

design + open note · steel twister · claude-code · design mode — models build from scratch

excitementintensity×build failed
0515intensity ceiling 10round 1: excitement 6.846.84round 2: excitement 6.79round 1: intensity 9.76round 2: intensity 8.46R1R2
score
6.84
intensity
9.76
nausea
5.67
best round
1
rated rounds
2/2
similarity
0.42
tokens / cost
165 in / 113k out · $6.55
2 round(s)

Round 1 trace (95, 2 rejected) →

excitement 6.84 intensity 9.76 nausea 5.67 length 611 drops 8 airtime 40 tested ok similarity 0.42 (nearest: Dizzymouse)
track program (97 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"
    },
    {
      "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_to_down_25",
    "right_turn_3_down_25",
    "right_turn_3_down_25",
    "right_turn_3_down_25",
    "right_turn_3_down_25",
    "down_25",
    "down_25",
    "down_25",
    "down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "half_loop_up",
    "right_corkscrew_down",
    "flat_to_down_25",
    "down_25",
    "down_25",
    "down_25",
    "down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "right_vertical_loop",
    "down_25_to_flat",
    "right_corkscrew_up",
    "left_corkscrew_down",
    "right_corkscrew_up",
    "left_corkscrew_down",
    "flat_to_left_bank",
    "banked_left_turn_5",
    "left_bank_to_flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "flat_to_down_25",
    "down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "flat_to_down_25",
    "down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "flat_to_down_25",
    "down_25",
    "down_25_to_flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat_to_left_bank",
    "banked_left_turn_5",
    "left_bank_to_flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat_to_left_bank",
    "banked_left_turn_5",
    "left_bank_to_flat"
  ],
  "ride_type": 51,
  "start": {
    "dir": 0,
    "x": 62,
    "y": 60
  }
}
park screenshot, claude-opus-5 round 1

Round 2 trace (47) →

excitement 6.79 intensity 8.46 nausea 4.38 length 747 drops 10 airtime 63 tested ok similarity 0.38 (nearest: Viper)
track program (122 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"
    },
    {
      "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"
    },
    {
      "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_to_left_bank",
    "banked_left_turn_5",
    "left_bank_to_flat",
    "flat_to_down_25",
    "down_25",
    "down_25_to_down_60",
    "down_60",
    "down_60",
    "down_60",
    "down_60_to_down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "half_loop_up",
    "right_corkscrew_down",
    "flat_to_down_25",
    "down_25",
    "down_25",
    "down_25",
    "down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "half_loop_up",
    "left_corkscrew_down",
    "flat_to_down_25",
    "down_25",
    "down_25",
    "down_25",
    "down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "right_vertical_loop",
    "down_25_to_flat",
    "flat_to_up_25",
    "left_vertical_loop",
    "down_25_to_flat",
    "flat_to_left_bank",
    "banked_left_turn_5",
    "left_bank_to_flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "flat_to_down_25",
    "down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "flat_to_down_25",
    "down_25",
    "down_25_to_flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat_to_left_bank",
    "banked_left_turn_5",
    "left_bank_to_flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "flat_to_down_25",
    "down_25",
    "down_25_to_flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "flat_to_down_25",
    "down_25",
    "down_25_to_flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat",
    "flat_to_left_bank",
    "banked_left_turn_5",
    "left_bank_to_flat",
    "flat_to_down_25",
    "down_25_to_flat",
    "flat",
    "flat",
    "brakes",
    "brakes"
  ],
  "ride_type": 51,
  "start": {
    "dir": 1,
    "x": 62,
    "y": 35
  }
}
park screenshot, claude-opus-5 round 2

← 20260725-opus5-python-probe · all runs