Finding
Opus 5.5 drew the cleanest French presses on the bench, but it put the filter stack the right way up in only one run of three. In the third run it had the order right in round 1, then decided the spiral plate goes at the bottom and swapped the two plates.
The prompt, v2, frozen
Draw an exploded assembly diagram of a French press, as an SVG. Use the standard three-piece plunger filter rather than a two-piece travel press.
Checks
| check · how it is decided | run 1 | run 2 | run 3 |
|---|---|---|---|
| All seven partsby eye · Pass if all seven are drawn as separate shapes and each has a label naming it. A part that is only named in a label or a comment, with nothing drawn, fails it. | ✓ pass | ✓ pass | ✓ pass |
| Stack ordermeasured · Knob, lid, spiral plate, mesh, cross-plate and carafe, top to bottom along the axis, read from each part's y coordinate in the SVG. Its comment is ignored. The rod is left out, because it runs through the lid. | ✗ fail | ✓ pass | ✗ fail |
| Filter fits the glassmeasured · Every filter piece drawn is 80% to 105% of the width of the carafe's glass wall, measured in the SVG. | ✓ pass | ✓ pass | ✗ fail |
| One assembly axisby eye · Pass if one vertical line runs through the middle of every part, and every leader line leaves from the same side. One part pushed off to the side fails it. | ✓ pass | ✗ fail | ✓ pass |
| Labels matchby eye · Pass if following each leader line from its label ends on the part that label names. One label pointing at the wrong part, or at nothing, fails it. | ✓ pass | ✓ pass | ✓ pass |
Ground truth: The standard three-piece plunger, from the top down: spiral (spring) plate, mesh screen, cross-plate. The cross-plate is the bottom piece, because its threaded stem is what the rod screws into. The spiral plate above the mesh is what presses the mesh edge out against the glass. Checked against the WebstaurantStore anatomy diagram, which is drawn exploded and labelled, and against each part's y coordinate in the SVG rather than its comment. Two-piece travel presses exist and stack differently, which is why v2 of the prompt names the design. Every model on this task.
Run 1 has the stack upside down. Run 2 has it right, but it draws the frame beside the glass, so it fails the axis check.
Run 3 first had the right order then its reasoning said "Actually the correct Bodum order is cross plate, mesh, then spiral plate at the bottom with the spring edge facing the glass", and it moved both plates with a translate. In its closing note it said it was not certain of the order and had not checked a manufacturer diagram. So it knew it was guessing, and it guessed wrong.
Failure types: part-order twice and physical-plausibility once.
Verdict
Verdict
It draws better than anything else here and knows the French press no better. Pareto and Sol each got the stack right in two runs of three.
n = 3 runs a model. p = 1.0 against Pareto or Sol, Fisher exact, after the runs.
The three runs cost $0.64 and took about 85 seconds each.