OverviewCode › R4
Code · archive round

R4 · Sorting race: bubble vs insertion vs quicksort

Archive round — this round predates the blind-first layout, so answers appear with model names attached. Newer rounds keep the models anonymous until you reveal them.
Scores on this page are the original editorial 0–10 ratings from the 2026-07-17 write-up (writing / art-direction / runtime quality) — not blind rubric scores against a documented gold answer, which is how R31 onward are scored. The underlying answers are unchanged from that write-up.
editorial scores · 2026-07-17 5/7 models answered
Show task
Three sorting algorithms — bubble, insertion, quick — racing side by side on the SAME shuffled 60-bar array. Animated, auto-starting, auto-restarting, live comparison counts, highlighted comparisons. NO user input.

Answers

🥇Claude Sonnet 5FRONTIEROK★ 9.2/10
RUNS ✓quick DONE @488 cmp while bubble grinds at 1404

Tells the algorithmic story perfectly: finish states, pivot highlights, glowing done-panels. Textbook.

🥈GPT-5.5 · CodexCLIOK★ 8.3/10
RUNS ✓steady progress, clean panels

Redemption after R3 — correct, brisk, minimal. Lacks finish-state drama.

🥉Claude Opus 4.8FRONTIEROK★ 8.0/10
RUNS ✓generator-based stepping

Elegant generator architecture, slightly leisurely pacing on screen.

4Claude Fable 5HOUSEOK★ 7.5/10
RUNS ✓horizontal lanes, value-hue bars

The only distinct layout of the round — but the hue-coded bars read noisy next to Sonnet's clarity.

5qwen3:14bLOCALISSUES★ 2.0/10
headers render, 0 comparisons, no bars

Valid syntax, dead runtime — the shell renders, nothing sorts.

6qwen3-coder:30bLOCALCRASH★ 1.5/10
SyntaxError — script never parses

17.8 KB of ambition, one parse error, zero bars.

7qwen3.5:9bLOCALCRASH★ 1.5/10
SyntaxError — script never parses

Same story, 22 KB heavier.