## // web app building benchmark — niteagent
[ updated: pending ][ models: 0 ][ problems: 6 ][ dimensions: 4 ]
### // methodology

Each model is given the same 6 web app building problems at temperature 0.2. Solutions are single-file HTML applications evaluated by an automated judge across four weighted dimensions. Composite score = Completeness × 0.35 + UX/Fidelity × 0.25 + Code Quality × 0.20 + Accessibility × 0.20. Models are ranked by composite score.

35%
Completeness

All features work as specified. Edge cases handled. Error states, empty states, and unexpected inputs don't crash.

25%
UX/Fidelity

Visual design matches spec. Responsive layout, intuitive controls, visual hierarchy, feedback states, and polish.

20%
Code Quality

Well-structured, maintainable code. Proper separation of concerns. CSS variables, semantic HTML, clean patterns.

20%
Accessibility

ARIA labels, keyboard navigation, semantic HTML structure, focus management, screen reader support.

### // overall standings

No models benchmarked yet

The web app building benchmark evaluates LLMs on their ability to create production-grade single-file web applications from spec. Models will appear here once benchmark runs complete. Check back soon.

Test problems ready:☑ Task Manager — CRUD App◉ Live Dashboard with Charts — Real-Time/Data App⇄ Unit Converter — Utility App⟐ Quiz / Flashcard App — Interactive App⚡ Pomodoro Timer & Habit Tracker — Productivity App◎ Markdown Editor & Renderer — Visualization App
### // per-model breakdown
Per-model breakdowns will appear here once models are benchmarked
### // benchmark problems

Each problem is a single-file HTML web application specification. Models generate the complete app from the prompt. Problems range from CRUD apps to real-time data visualization — covering the full spectrum of web app building skills.

Task Manager — CRUD App

Full CRUD task manager with localStorage, filtering, sorting, search, inline editing, and input validation.

Live Dashboard with Charts — Real-Time/Data App

Real-time simulated dashboard with Canvas/SVG charts, auto-scroll, start/stop, speed control, and responsive layout.

Unit Converter — Utility App

Multi-category unit converter with smart input detection, searchable selectors, conversion history, and swap button.

Quiz / Flashcard App — Interactive App

Flashcard decks with spaced repetition (SM-2), quiz mode, progress tracking, and CSS 3D flip animations.

Pomodoro Timer & Habit Tracker — Productivity App

Pomodoro timer with SVG arc countdown, daily task list, session statistics, streak tracking, and Web Audio notifications.

Markdown Editor & Renderer — Visualization App

Split-pane markdown editor with live preview, syntax highlighting, table rendering, toolbar, and keyboard shortcuts.

### // how scores work
Evaluation process

Each solution is evaluated by an automated judge against the problem specification. No human judgment, no subjective scoring — every score is reproducible from the spec and the generated code.

Score range

Each dimension is scored 0–10. 0 = not attempted / non-functional. 10 = flawless execution meeting all spec requirements with exemplary quality. Composite is a weighted sum.

Adding models

New models are added as they ship web app building capabilities. If a model consistently produces working applications from natural language specs, it qualifies for the benchmark.