Compare N assets

# one-time: the browser this npx CLI renders with
npx -p @spatialpack/cli playwright install chromium

for a in *.glb; do
  for b in *.glb; do
    [[ "$a" < "$b" ]] || continue
    npx @spatialpack/cli visual-diff "$a" "$b" \
      --grid orbit-8 --out "diff/${a%.glb}__${b%.glb}"
  done
done

One visual-diff per pair, upper-triangle so no pair runs twice — 3 runs for 3 files, 28 for 8. Each writes its own visual-diff.json with SSIM + ΔE94 across the orbit-8 grid — the same grid the published benchmark uses. Those are the cells of the matrix; no CLI command draws the grid itself, and assembling the cells into that view is the part we would have done for you.

Pass --grid explicitly: it defaults to a 6-angle mixed-pitch grid, which scores differently. The published leaderboard scored one view of orbit-8 (--views 1); the default here is the stricter worst-of-8. The first line is once per machine — npx gives the CLI its own Playwright, and its browser is not the one any other install has. Free and unmetered.

- uses: ./actions/spatialpack-action
  with:
    visual-diff: true

The GitHub Action runs the diff on your own runner and fails the build past your ΔE94 threshold. Also free.