Test results

Pick a model to see how accessible its components are, with the accessibility skill off and on.

GPT-6 Luna

Skill version 2.3

Totals for the whole run, across 3 generations of each of the 22 components
Measure	Skill off	Skill on
Failed acceptance criteria	143	12
of which WCAG	100	7
of which Best practice	43	5
Total cost (US dollars)	$0.09	$0.36
Input tokens	199,442	6.2 million
Output tokens	132,160	369,858

Button

Results, totals across all generations

Cost in USD

Failed acceptance criteria

Skill off

3 fails

WCAG: 3 fails
Best practice: 0 fails

Skill on

0 fails

WCAG: 0 fails
Best practice: 0 fails

Total cost

Skill off: $0.0036
Skill on: $0.01

Efficiency, average per generation

Input tokens include cache reads.

Input tokens

Skill off: 2,987 tokens
Skill on: 79,864 tokens

Output tokens

Skill off: 1,777 tokens
Skill on: 3,463 tokens

Turns

Skill off: 3 turns
Skill on: 8 turns

Time

Skill off: 17 seconds
Skill on: 42 seconds

Cost

Skill off: $0.0012
Skill on: $0.0040

Prompt given to the model

Build a Button component in a single self-contained HTML file (inline CSS and JS, no build step, no external libraries). Include two buttons on the page: one enabled button labelled "Save changes" and one disabled button labelled "Delete". Output the complete code in one HTML code block.

Generated code

Skill off: Generation 1, skill off (opens in a new tab)
Generation 2, skill off (opens in a new tab)
Generation 3, skill off (opens in a new tab)
Skill on: Generation 1, skill on (opens in a new tab)
Generation 2, skill on (opens in a new tab)
Generation 3, skill on (opens in a new tab)

Showing 12 of 12 criteria

Labels and messaging1 criterion

The button has a visible label.WCAG 3.3.2Automated
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass

Semantic markup5 criteria

The button's role is included in the accessibility tree.WCAG 4.1.2Automated
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
The button's accessible name is included in the accessibility tree.WCAG 4.1.2Automated
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
The button's accessible name contains the text of its visible label.WCAG 2.5.3Automated
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
The button's accessible name matches its visible label exactly, or at least starts with the exact text of the visible label.Best practiceAutomated
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
If the button is disabled, the button's disabled state is included in the accessibility tree.WCAG 4.1.2Automated
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass

Keyboard4 criteria

The button is focusable using a keyboard.WCAG 2.1.1Browser
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
The button is activated by pressing the Space or Enter keys on the keyboard.Best practiceBrowser
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
The button has a focus style when it receives focus using a keyboard.WCAG 2.4.7Browser
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
The keyboard focus style still renders in forced-colours mode (e.g. Windows High Contrast Mode).Best practiceBrowser
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass

Visual design2 criteria

All text meets the minimum contrast ratio of 4.5:1 against the background colour, or 3:1 for large-scale text.WCAG 1.4.3Automated
Skill off
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass
If the button uses a custom focus style, the focus indicator has a contrast ratio of at least 3:1 against its background colour.WCAG 1.4.11Browser
Skill off
3/3 failed
- Generation 1: Fail
- Generation 2: Fail
- Generation 3: Fail
Skill on
3/3 passed
- Generation 1: Pass
- Generation 2: Pass
- Generation 3: Pass