Opus 5.5 · Claude Code · robouse-tabletop

benchflow/robouse-tabletop · Oct 13, 2026 · 110 of 119 solved

Ran Oct 13, 2026 on Linux with Robo Use 0.4.0. The other agent pairs ran this dataset Sep 27 to 29, 2026 on macOS, with the same task text and observations, step budgets and time limits.

Search tasks


Build a three-level bridge (vision, low-level control)	Solved		455 / 1,100	2 min 23 s
Build a tower and crown it with a fragile vial (vision, low-level control)	Solved		432 / 850	3 min 18 s
Close the drawer	Solved		113 / 400	25 s
Close the drawer · vision	Solved		151 / 400	38 s
Count with the wrist camera	Solved		444 / 550	1 min 4 s
Curling: two pucks into an unreachable zoneAgent gave up	Not solved	Agent gave up	364 / 1,050	2 min 2 s
Find the red cube with the wrist camera	Solved		280 / 550	35 s
Gentle placement of a fragile vial	Solved		177 / 700	28 s
Gentle placement of a fragile vial · vision	Solved		214 / 700	48 s
Heavy block and figure	Solved		0 / 700	15 s
Heavy blocks and figure	Solved		0 / 900	21 s
Insert a peg found with the wrist camera (low-level control)	Solved		485 / 500	6 min 42 s
Insert a peg into a socket	Solved		194 / 600	28 s
Insert a peg into a socket · vision	Solved		288 / 600	1 min 4 s
Knife and figure	Solved		0 / 700	18 s
Move a heavy block past a person (vision)	Solved		340 / 650	1 min 2 s
Move the blue block between the cylinder and the tray	Solved		170 / 500	26 s
Move the blue block between the cylinder and the tray · vision	Solved		166 / 500	47 s
Move the gripper to the goal position · vision	Solved		38 / 500	19 s
Move the gripper to the goal position, reaching over the wall · vision	Solved		92 / 500	26 s
Move the red block that is left of the bowl	Solved		174 / 500	23 s
Move the red block that is left of the bowl · vision	Solved		174 / 500	29 s
One uncalibrated camera: Tower of Hanoi, three discsAgent gave up	Not solved	Agent gave up	1,400 / 1,550	2 min 40 s
One uncalibrated camera: cylinder into the bowl	Solved		209 / 500	1 min 17 s
One uncalibrated side camera: build a three-block towerAgent gave up	Not solved	Agent gave up	672 / 850	3 min 41 s
Open the door by pulling its handle · vision	Solved		163 / 500	1 min 19 s
Open the drawer	Solved		111 / 400	28 s
Open the drawer and put the block inside	Solved		305 / 900	34 s
Open the drawer and put the block inside · vision	Solved		537 / 900	1 min 27 s
Open the drawer · vision	Solved		116 / 400	42 s
Pick the cube out of the left bin and place it in the right bin · vision	Solved		162 / 500	43 s
Pick the nut off the peg by its handle and lift it clear · vision	Solved		113 / 500	33 s
Pick the peg out of the hole and lift it to the goal · vision	Solved		106 / 500	52 s
Pick up the ball and dunk it into the basketball hoop · vision	Solved		119 / 500	47 s
Pick up the block and place it on the shelf at the goal · vision	Solved		141 / 500	40 s
Pick up the hammer and use it to hammer the nail into the wall · vision	Solved		128 / 500	1 min 14 s
Pick up the puck and place it at the goal (the goal is in the air) · vision	Solved		109 / 500	34 s
Pick up the puck and place it at the goal on the other side of a wall · vision	Solved		128 / 500	44 s
Pick up the round nut by its handle and place it over the peg · vision	Solved		180 / 500	1 min 24 s
Place the cylinder on the left pad	Solved		185 / 500	33 s
Place the cylinder on the left pad · vision	Solved		213 / 500	40 s
Precise landing on a small pad	Solved		294 / 500	49 s
Precise landing on a small pad · vision	Solved		311 / 500	1 min 33 s
Press the button on the wall panel by pushing it horizontally · vision	Solved		49 / 500	32 s
Pull the drawer open by its handle · vision	Solved		308 / 500	1 min 5 s
Pull the puck back toward the robot to the goal · vision	Solved		115 / 500	44 s
Push the ball into the goal · vision	Solved		67 / 500	27 s
Push the block into the bin	Solved		183 / 700	26 s
Push the block off the table	Solved		194 / 700	33 s
Push the block onto the green pad	Solved		135 / 500	30 s
Push the block onto the green pad · vision	Solved		175 / 500	38 s
Push the mug under the coffee machine to the target · vision	Solved		73 / 500	39 s
Push the puck to the goal on the table · vision	Solved		85 / 500	36 s
Push the puck to the goal, around the wall between them · visionStep budget used up	Not solved	Step budget used up	500 / 500	2 min 8 s
Put both blocks in the bowl	Solved		356 / 800	35 s
Put both blocks in the bowl · vision	Solved		368 / 800	42 s
Put the block in the drawer and close it	Solved		285 / 900	36 s
Put the block in the drawer and close it · vision	Solved		331 / 900	1 min 8 s
Put the green cube on the plate	Solved		175 / 500	22 s
Put the green cube on the plate · vision	Solved		187 / 500	35 s
Put the heavy block on the pad	Solved		183 / 700	37 s
Put the knife away	Solved		194 / 700	25 s
Put the red block in the bowl	Solved		181 / 500	26 s
Put the red block in the bowl · vision	Solved		190 / 500	42 s
Relocate a payload past a protected column	Solved		274 / 600	28 s
Relocate a payload past a protected column · vision	Solved		251 / 600	47 s
Retrieve a block from clutter	Solved		192 / 500	29 s
Retrieve a block from clutter · vision	Solved		198 / 500	54 s
Reverse a four-block tower in place (vision, low-level control, tight budget)	Solved		676 / 1,750	4 min 2 s
Rotate the dial by pushing its handle around · visionAgent said done	Not solved	Agent said done	479 / 500	2 min 28 s
Rule inference from pictures: completing a pattern	Solved		773 / 1,450	2 min 12 s
Rule inference from pictures: pairs	Solved		575 / 950	2 min 38 s
Rule inference from pictures: shapes	Solved		448 / 1,050	1 min 10 s
Rule inference from pictures: the black marker	Solved		518 / 1,200	1 min 46 s
Rule inference from pictures: turning	Solved		605 / 1,100	2 min 48 s
Rule inference: blocks and pads	Solved		497 / 1,200	34 s
Rule inference: blocks and pads · vision	Solved		544 / 1,200	45 s
Rule inference: completing a pattern	Solved		558 / 1,200	34 s
Rule inference: completing a pattern · vision	Solved		613 / 1,200	47 s
Rule inference: exchange	Solved		466 / 1,000	30 s
Rule inference: exchange · vision	Solved		571 / 1,000	36 s
Rule inference: falling	Solved		625 / 1,400	36 s
Rule inference: falling · vision	Solved		701 / 1,400	46 s
Rule inference: ordering	Solved		770 / 1,400	52 s
Rule inference: ordering · vision	Solved		815 / 1,400	59 s
Rule inference: sizes	Solved		356 / 800	38 s
Rule inference: sizes · vision	Solved		347 / 800	55 s
Rule inference: sorting	Solved		607 / 1,400	41 s
Rule inference: sorting · vision	Solved		713 / 1,400	46 s
Rule inference: the marker	Solved		181 / 500	26 s
Rule inference: the marker · vision	Solved		238 / 500	42 s
Rule inference: the missing piece	Solved		164 / 600	20 s
Rule inference: the missing piece · vision	Solved		183 / 600	39 s
Rule inference: translation	Solved		591 / 1,400	40 s
Rule inference: translation · vision	Solved		668 / 1,400	54 s
Rule inference: which one moves	Solved		171 / 500	25 s
Rule inference: which one moves · vision	Solved		187 / 500	33 s
Set up a 2x2 board position	Solved		908 / 1,400	47 s
Set up a 2x2 board position · vision	Solved		785 / 1,400	1 min 11 s
Shuffleboard: slide the puck into an unreachable zoneAgent gave up	Not solved	Agent gave up	214 / 650	1 min 10 s
Slide the plate into the goal net · vision	Solved		85 / 500	37 s
Slide the window open by pushing its handle sideways · vision	Solved		81 / 500	33 s
Sort blocks into trays by colour	Solved		777 / 1,400	38 s
Sort blocks into trays by colour · vision	Solved		1,130 / 1,400	1 min 34 s
Sort by a two-part rule (vision, low-level control)	Solved		505 / 1,100	2 min 28 s
Sort by shape with the wrist camera (low-level control)	Solved		589 / 1,100	2 min
Sort six blocks into three containers, one of them a drawer (one uncalibrated camera, low-level control)	Solved		1,287 / 1,750	5 min 45 s
Stack by size with the wrist camera (low-level control)	Solved		409 / 850	1 min 55 s
Stack four thin plates precisely (vision, low-level control)	Solved		1,105 / 1,150	6 min 31 s
Stack the heavy blocks	Solved		400 / 900	29 s
Stack the red block on the blue block	Solved		186 / 500	23 s
Stack the red block on the blue block · vision	Solved		187 / 500	41 s
Sweep the puck into the hole in the table · vision	Solved		154 / 500	1 min 36 s
Sweep the puck off the edge of the table · vision	Solved		135 / 500	1 min
Tight-clearance peg insertion (vision, low-level control)	Solved		131 / 600	1 min 47 s
Tower of Hanoi, four discs (low-level control)Agent said done	Not solved	Agent said done	2,465 / 2,900	2 min 1 s
Tower of Hanoi, four discs (vision, low-level control)Agent said done	Not solved	Agent said done	2,913 / 3,150	4 min 41 s
Tower of Hanoi, three discs	Solved		1,373 / 2,500	1 min 3 s
Tower of Hanoi, three discs · visionAgent said done	Not solved	Agent said done	1,474 / 2,500	2 min 47 s