← 2DBench

Sketch Island Bench

3 models · 2026-10-30
Recreate this image (/work/ref/sketch.jpg) as a scene in three.js, as closely as you can.
Opus 5.5 output▶ View live
anthropic logo
Opus 5.5high
subtodo run, original prompt pointed at the full-res PNG
$1.7113m 32s5 fpslive
GPT 6.1 Sol output▶ View live
openai logo
GPT 6.1 Solhigh
subtodo run, original prompt pointed at the full-res PNG
$0.094m 51s60 fpslive
Kimi K3.1 Pro output▶ View live
moonshotai logo
Kimi K3.1 Pro
subtodo run, original prompt pointed at the full-res PNG
$0.6518m 19s60 fpslive
Each model runs in its own sandbox through the TODO for AI agent — identical prompt, no view of the others. We publish the raw outputs, what they cost and how long they took. No Elo, nothing fitted to a curve. Upvotes are a separate reader signal and never change the order. Reproducible from the benchmarks repo.