- steps
- 17
- duration
- 1m 20s
- errors
- 0
- files
- 5
session rating0
Build a unit converter
Build a unit converter as a static web app in this directory: index.html, style.css, convert.js. Categories: length, mass, temperature, data size, time. Instant two-way conversion between any pair in a category, swap button, precision selector, recent-conversions list in localStorage, keyboard-first UX, dark theme, and tests.html + tests.js exercising the conversion math including temperature edge cases. Run the tests headless with node and fix what fails.
3 sessions from the same published task, on one clock. Scrub to any moment and compare what existed, or open a pane to read its full trajectory.
0 votes · scrub before you decide
file writeshellreasoningerror / retrymessageread / searchsubagentplantool call
- steps
- 26
- duration
- 2m 38s
- errors
- 1
- files
- 5
session rating0
- steps
- 24
- duration
- 3m 36s
- errors
- 0
- files
- 5
session rating0
| gpt-5.5 · low | gpt-5.5 · medium | gpt-5.5 · high | |
|---|---|---|---|
| steps | 17 | 26 | 24 |
| tool calls | 7 | 9 | 8 |
| duration | 1m 20s | 2m 38s | 3m 36s |
| tokens out | 1k | 3k | 5k |
| reasoning | ~7 | ~63 | ~129 |
| compactions | 0 | 0 | 0 |
| subagents | 0 | 0 | 0 |
| errors | 0 | 1 | 0 |
| files | 5 | 5 | 5 |
Token and context figures come from the harness's own per-turn reporting; sessions captured before telemetry landed show a dash. Reasoning marked ~ is estimated from the transcript's thinking text where the harness doesn't report it separately.
Verdict
Pairwise picks on the outputs above, from the current Arena tally.
What voters said they were judging
Arena picks do not collect a reason or judging criterion, so no qualitative verdict is inferred here. The bars report only the current comparison tally.