SVG: exploded French press, by command-a-plus
command-a-plus · round 1 · seed 3$0.02SVGPNGpress.svg
press.svg · 41 lines
<?xml version="1.0" encoding="UTF-8"?>
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 600 400" width="600" height="400">
  <defs>
    <pattern id="meshPattern" patternUnits="userSpaceOnUse" width="10" height="10">
      <path d="M0 10 L10 0 M-5 5 L5 -5 M5 5 L-5 -5" stroke="#ccc" stroke-width="0.5" fill="none"/>
    </pattern>
  </defs>

  <!-- Glass beaker -->
  <rect x="50" y="50" width="120" height="300" fill="#e6f3ff" stroke="#333" stroke-width="2"/>
  <!-- Lid -->
  <rect x="50" y="10" width="120" height="30" fill="#606060" stroke="#333" stroke-width="2"/>
  <!-- Plunger rod -->
  <rect x="200" y="80" width="20" height="250" fill="#a0a0a0" stroke="#333" stroke-width="2"/>
  <!-- Filter top piece -->
  <rect x="230" y="90" width="80" height="20" fill="#d0d0d0" stroke="#333" stroke-width="2"/>
  <!-- Filter mesh -->
  <rect x="230" y="130" width="80" height="20" fill="url(#meshPattern)" stroke="#333" stroke-width="2"/>
  <!-- Filter bottom piece -->
  <rect x="230" y="250" width="80" height="20" fill="#c0c0c0" stroke="#333" stroke-width="2"/>

  <!-- Assembly guide lines (dashed) -->
  <g stroke="#999" stroke-dasharray="5,5" stroke-width="1">
    <!-- Lid alignment -->
    <line x1="50" y1="40" x2="170" y2="40"/>
    <!-- Plunger insertion into glass -->
    <line x1="210" y1="80" x2="210" y2="330"/>
    <!-- Filter pieces onto plunger -->
    <line x1="230" y1="90" x2="210" y2="90"/>
    <line x1="230" y1="130" x2="210" y2="130"/>
    <line x1="230" y1="250" x2="210" y2="250"/>
  </g>

  <!-- Labels -->
  <text x="85" y="20" font-size="12" fill="#333" text-anchor="middle">Lid</text>
  <text x="85" y="370" font-size="12" fill="#333" text-anchor="middle">Glass Beaker</text>
  <text x="210" y="200" font-size="12" fill="#333" text-anchor="middle">Plunger Rod</text>
  <text x="270" y="100" font-size="10" fill="#333" text-anchor="middle">Filter Top</text>
  <text x="270" y="140" font-size="10" fill="#333" text-anchor="middle">Filter Mesh</text>
  <text x="270" y="260" font-size="10" fill="#333" text-anchor="middle">Filter Bottom</text>
</svg>
gallery/Exploded French press/Command A+, run 3

Terrible

modelcohere/command-a-plustasksvg-french-press@v2date2026-09-27rounds1 of 8 allowedcost$0.0209thinkingmediumtokens7,160 in, 12,526 out, of which 11,588 reasoningtime1 min 3 sfailurecompleteness · physical-plausibility · labelling
All seven partsdecided by looking · 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.fail
Stack orderdecided by a number, from the filefail
Filter fits the glassdecided by a number, from the filefail
One assembly axisdecided by looking · 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.fail
Labels matchdecided by looking · 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.fail

There is no post about this one, since only some results need one, so the picture, the source and the scoring are here instead.

its other runs
the prompt it was given
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.
Frozen at v2. This model got this and nothing else.
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. How it is scored, and this run's files on GitHub.