PREVIEW · the site is new and still changing
Nerdict
Claude Haiku 5.5highin Claude CodeBrowser OS
Score19/19all 19 checks passed
Cost$0.060
Time7m21s
Runs of this setup#1 19/19#2 19/19#3 19/19
Browser OS by Claude Haiku 5.5Open it running ↗

This is the exact file the model wrote, unedited, served from an isolated domain with the network blocked, as in the test.

What the robot did

The comparison clip (the same script for every model)

Checks

19/19

Written from the prompt (test v1, checks v4). The model never saw them.

  • ✓Opens without a page error
  • ✓Taskbar and clock visible10:34:38
  • ✓Animated wallpaperquadros diferentes
  • ✓Right-click menu on the desktop5 elemento(s) novo(s) perto do clique
  • ✓Double-click opens Notes
  • ✓Double-click opens Calculator
  • ✓Double-click opens Files
  • ✓Double-click opens Game
  • ✓Double-click opens Settings
  • ✓Window drags by its title bar(110,60) → (350,210)
  • ✓Maximize fills the screen1920×1032
  • ✓Minimize hides the windowescondida
  • ✓Close removes the windowsumiu
  • ✓Clicking a window brings it to the frontapós clicar nas notas: notes; após clicar na calculadora: calculator
  • ✓Calculator: 12 + 30 = 42visor: '42'
  • ✓Notes survive a reloaddepois do reload: 'Welcome to Notes.\n\nnerdict-7319 hello from the Nerdict bench'
  • ✓No uncaught exception during the session
  • ✓No console errors
  • ✓No network attempts
Model
Claude Haiku 5.5 claude-haiku-5-5
Reasoning effort
high
Started
Oct 10, 00:15 UTC
Harness
Claude Code 2.1.296, by Anthropic (not ours) · one shot · tools used: 0
Harness settings
print mode, safe mode, every tool off (--tools ""), no saved session, empty folder outside Git · CLAUDE_CODE_MAX_OUTPUT_TOKENS=128000
Tokens
in 2 · cache read 3,189 · out 118,795 · reasoning 88,904 · requests 1
Cost
$0.060 API-equivalent · extra spend $0 (subscription)
Test
Browser OS v1 · checks v4 · prompt 213e11b5fdb9
Output file
63.1 KB · sha256 b063ed251559
Run id
20261010T001509Z_browser-os-v1_haiku-high_r1