CLAUDE-SONNET-5 · 20260723-v3-twister-design

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

claude-sonnet-5 · 3rd of 3 compare all models →

design · steel twister · driver-api · design mode — models build from scratch

excitementintensity×build failed
0515intensity ceiling 10×round 1: build failed (40/40 pieces placed): set_status(testing): Track is not a complete circuit [track starts at tile (60, 60, z=112, dir=0, bank=0, slope=0) and ends at tile (50, 76, z=176, dir=0, bank=0, slope=0); a closed circuit requires these to be identical]R1×round 2: build failed (37/37 pieces placed): set_status(testing): Track is not a complete circuit [track starts at tile (60, 60, z=112, dir=0, bank=0, slope=0) and ends at tile (51, 76, z=176, dir=0, bank=0, slope=0); a closed circuit requires these to be identical]R2×round 3: build failed at piece 38 (38/41 pieces placed): rejected at cursor (x=53, y=71, z=160, dir=3): Twister Roller Coaster 1 in the wayR3×round 4: build failed at piece 37 (37/38 pieces placed): rejected at cursor (x=50, y=69, z=176, dir=2): Twister Roller Coaster 1 in the wayR4×round 5: build failed (37/37 pieces placed): set_status(testing): Track is not a complete circuit [track starts at tile (60, 60, z=112, dir=0, bank=0, slope=0) and ends at tile (51, 74, z=176, dir=0, bank=0, slope=0); a closed circuit requires these to be identical]R5×round 6: build failed (38/38 pieces placed): set_status(testing): Track is not a complete circuit [track starts at tile (60, 60, z=112, dir=0, bank=0, slope=0) and ends at tile (50, 70, z=176, dir=2, bank=0, slope=0); a closed circuit requires these to be identical]R6
score
—
intensity
—
nausea
—
best round
—
rated rounds
0/6
tokens / cost
—
6 round(s)

Round 1 build failed

build failed (40/40 pieces placed): set_status(testing): Track is not a complete circuit [track starts at tile (60, 60, z=112, dir=0, bank=0, slope=0) and ends at tile (50, 76, z=176, dir=0, bank=0, slope=0); a closed circuit requires these to be identical]

track program (40 pieces)
{
  "pieces": [
    "begin_station",
    "middle_station",
    "end_station",
    {
      "t": "flat_to_up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    "up_25_to_up_60",
    {
      "t": "up_60"
    },
    "up_60_to_flat",
    "flat",
    "left_turn_3",
    "left_turn_3",
    "flat_to_down_25",
    "down_25",
    "down_25_to_down_60",
    "down_60",
    "down_60_to_flat",
    "flat",
    "flat_to_up_25",
    "left_vertical_loop",
    "down_25_to_flat",
    "flat",
    "left_turn_5",
    "left_turn_5",
    "flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "flat",
    "right_turn_5",
    "left_corkscrew_up",
    "right_corkscrew_down",
    "right_turn_5",
    "flat",
    "left_turn_5",
    "left_turn_5",
    "flat",
    "brakes",
    "flat"
  ],
  "ride_type": 51,
  "start": {
    "dir": 0,
    "x": 60,
    "y": 60
  }
}
park screenshot, claude-sonnet-5 round 1 view 1

Round 2 build failed

build failed (37/37 pieces placed): set_status(testing): Track is not a complete circuit [track starts at tile (60, 60, z=112, dir=0, bank=0, slope=0) and ends at tile (51, 76, z=176, dir=0, bank=0, slope=0); a closed circuit requires these to be identical]

track program (37 pieces)
{
  "pieces": [
    "begin_station",
    "middle_station",
    "end_station",
    {
      "t": "flat_to_up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    "up_25_to_up_60",
    {
      "t": "up_60"
    },
    "up_60_to_flat",
    "flat",
    "left_turn_3",
    "left_turn_3",
    "flat_to_down_25",
    "down_25",
    "down_25_to_down_60",
    "down_60",
    "down_60_to_flat",
    "flat",
    "flat_to_up_25",
    "left_vertical_loop",
    "down_25_to_flat",
    "flat",
    "left_turn_5",
    "left_turn_5",
    "flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "right_turn_5",
    "left_corkscrew_up",
    "right_corkscrew_down",
    "right_turn_5",
    "left_turn_5",
    "left_turn_5",
    "brakes",
    "flat"
  ],
  "ride_type": 51,
  "start": {
    "dir": 0,
    "x": 60,
    "y": 60
  }
}
park screenshot, claude-sonnet-5 round 2 view 1

Round 3 build failed

build failed at piece 38 (38/41 pieces placed): rejected at cursor (x=53, y=71, z=160, dir=3): Twister Roller Coaster 1 in the way

track program (41 pieces)
{
  "pieces": [
    "begin_station",
    "middle_station",
    "end_station",
    {
      "t": "flat_to_up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    "up_25_to_up_60",
    {
      "t": "up_60"
    },
    "up_60_to_flat",
    "flat",
    "left_turn_3",
    "left_turn_3",
    "flat_to_down_25",
    "down_25",
    "down_25_to_down_60",
    "down_60",
    "down_60_to_flat",
    "flat",
    "flat_to_up_25",
    "left_vertical_loop",
    "down_25_to_flat",
    "flat",
    "left_turn_5",
    "left_turn_5",
    "flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "right_turn_5",
    "left_corkscrew_up",
    "right_corkscrew_down",
    "right_turn_5",
    "left_turn_5",
    "left_turn_3",
    "left_turn_3",
    "flat_to_down_25",
    "down_25_to_flat",
    "brakes",
    "left_turn_3",
    "left_turn_3"
  ],
  "ride_type": 51,
  "start": {
    "dir": 0,
    "x": 60,
    "y": 60
  }
}
park screenshot, claude-sonnet-5 round 3 view 1

Round 4 build failed

build failed at piece 37 (37/38 pieces placed): rejected at cursor (x=50, y=69, z=176, dir=2): Twister Roller Coaster 1 in the way

track program (38 pieces)
{
  "pieces": [
    "begin_station",
    "middle_station",
    "end_station",
    {
      "t": "flat_to_up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    "up_25_to_up_60",
    {
      "t": "up_60"
    },
    "up_60_to_flat",
    "flat",
    "left_turn_3",
    "left_turn_3",
    "flat_to_down_25",
    "down_25",
    "down_25_to_down_60",
    "down_60",
    "down_60_to_flat",
    "flat",
    "flat_to_up_25",
    "left_vertical_loop",
    "down_25_to_flat",
    "flat",
    "left_turn_5",
    "left_turn_5",
    "flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "right_turn_5",
    "left_corkscrew_up",
    "right_corkscrew_down",
    "brakes",
    "flat",
    "left_turn_3",
    "left_turn_3",
    "left_turn_3",
    "left_turn_3"
  ],
  "ride_type": 51,
  "start": {
    "dir": 0,
    "x": 60,
    "y": 60
  }
}
park screenshot, claude-sonnet-5 round 4 view 1

Round 5 build failed

build failed (37/37 pieces placed): set_status(testing): Track is not a complete circuit [track starts at tile (60, 60, z=112, dir=0, bank=0, slope=0) and ends at tile (51, 74, z=176, dir=0, bank=0, slope=0); a closed circuit requires these to be identical]

track program (37 pieces)
{
  "pieces": [
    "begin_station",
    "middle_station",
    "end_station",
    {
      "t": "flat_to_up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    "up_25_to_up_60",
    {
      "t": "up_60"
    },
    "up_60_to_flat",
    "flat",
    "left_turn_3",
    "left_turn_3",
    "flat_to_down_25",
    "down_25",
    "down_25_to_down_60",
    "down_60",
    "down_60_to_flat",
    "flat",
    "flat_to_up_25",
    "left_vertical_loop",
    "down_25_to_flat",
    "flat",
    "left_turn_5",
    "left_turn_5",
    "flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "right_turn_3",
    "left_corkscrew_up",
    "right_corkscrew_down",
    "right_turn_3",
    "left_turn_5",
    "left_turn_5",
    "brakes",
    "flat"
  ],
  "ride_type": 51,
  "start": {
    "dir": 0,
    "x": 60,
    "y": 60
  }
}
park screenshot, claude-sonnet-5 round 5 view 1

Round 6 build failed

build failed (38/38 pieces placed): set_status(testing): Track is not a complete circuit [track starts at tile (60, 60, z=112, dir=0, bank=0, slope=0) and ends at tile (50, 70, z=176, dir=2, bank=0, slope=0); a closed circuit requires these to be identical]

track program (38 pieces)
{
  "pieces": [
    "begin_station",
    "middle_station",
    "end_station",
    {
      "t": "flat_to_up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    {
      "chain": true,
      "t": "up_25"
    },
    "up_25_to_up_60",
    {
      "t": "up_60"
    },
    "up_60_to_flat",
    "flat",
    "left_turn_3",
    "left_turn_3",
    "flat_to_down_25",
    "down_25",
    "down_25_to_down_60",
    "down_60",
    "down_60_to_flat",
    "flat",
    "flat_to_up_25",
    "left_vertical_loop",
    "down_25_to_flat",
    "flat",
    "left_turn_5",
    "left_turn_5",
    "flat",
    "flat_to_up_25",
    "up_25",
    "up_25_to_flat",
    "right_turn_3",
    "left_corkscrew_up",
    "right_corkscrew_down",
    "right_turn_3",
    "left_turn_1",
    "brakes",
    "left_turn_5",
    "left_turn_5",
    "left_turn_1"
  ],
  "ride_type": 51,
  "start": {
    "dir": 0,
    "x": 60,
    "y": 60
  }
}
park screenshot, claude-sonnet-5 round 6 view 1

← 20260723-v3-twister-design · all runs