Every answer on this page was written by an AI model, named on the answer. Answers can be wrong, out of date or made up, even when they sound sure. Check anything that matters before you rely on it.
On building web pages, two of the 13 models and settings scored 98 out of 100: Sonnet 5.5 at low effort and Fable 5.1 at low effort. Two more were within 5 points: Fable 5.1 at max effort (95) and Opus 5 at max effort (93). The lowest was Haiku 4.5, 63.
Written by an AI model from these results; every figure on this page is filled in by a program. How it was made.
Published
Score on building web pages
Each model and setting, highest first. A score here is the mean share of checks passed over the five prompts in this kind of work.
Score on building web pages
Light bars are low effort. Dark bars are max effort, or the only setting.
Show the numbers.
Score out of 100 on building web pages, for each model and setting, highest first
Model
Score
Sonnet 5.5 · low
98
Fable 5.1 · low
98
Fable 5.1 · max
95
Opus 5 · max
93
Fable 5 · max
93
Fable 5 · low
92
Opus 5.5 · low
90
Opus 5 · low
90
Sonnet 5.5 · max
80
Opus 5.5 · max
80
Sonnet 5 · low
79
Sonnet 5 · max
74
Haiku 4.5
63
Each prompt
Every cell is the share of that prompt’s checks the answer passed.
Score for each prompt in building web pages
Full marks
80 to 99
50 to 79
1 to 49
No checks passed
† The answer ran out of room: it reached the model’s output limit, mostly spent thinking, before it was finished. It is scored on what it did write.
Show the numbers.
Score out of 100 on each prompt in building web pages, for each model and setting
Prompt
Sonnet 5.5 low
Sonnet 5.5 max
Opus 5.5 low
Opus 5.5 max
Fable 5.1 low
Fable 5.1 max
Opus 5 low
Opus 5 max
Sonnet 5 low
Sonnet 5 max
Fable 5 low
Fable 5 max
Haiku 4.5
B01 A landing page for a small invoicing app
100
100
100
100
89
100
100
100
78
100
100
78
67
B02 Build a dashboard from a picture of its design
Asks for a one-page landing site for a small invoicing app for freelancers, as a single HTML file with no libraries and nothing loaded from outside. The page needs a headline, three named features, three price plans with a monthly and yearly switch (yearly gives two months free and is quoted as one total for the year), a four-question FAQ that opens and closes, and a sign-up button. It has to read well on a phone.
Every check passed: by 9 of 13 answers.
Missed most often:One h1, no skipped heading levels, text 16 px or larger, enough contrast, in 4 of 13 answers.
Sends a picture of a designer's mockup, an admin dashboard for a small bike repair shop, and asks for it to be built as one HTML file that matches the layout, colours, spacing and text as closely as it can. The bar chart has to be drawn to scale. On a phone the page must stack into one column, hide the sidebar behind a menu button and never scroll sideways.
Every check passed: by 5 of 13 answers.
Missed most often:Side by side with the picture, a designer would accept it as a faithful build, in 5 of 13 answers.
Asks for a personal kanban board in one HTML file with three columns: to do, doing and done. Cards are added by typing a title, can be dragged between and within columns, edited and deleted, and each column shows how many cards it holds. The board must survive a reload, work from the keyboard with the arrow keys, show titles that contain code exactly as typed, and fit a phone screen.
Every check passed: by 5 of 13 answers.
Missed most often:Dragging moves cards between columns and reorders them, and it survives a reload, in 5 of 13 answers.
Asks for the game Minesweeper in one HTML file: a 9 by 9 board with 10 mines, flags on right click, a mines-left counter, a first square that is never a mine, empty areas that open by themselves, win and lose messages and a new game button. Every square has to be a button so the game works from the keyboard, and the board has to fit a phone.
Every check passed: by 6 of 13 answers.
Missed most often:Numbers are right and an empty square opens its neighbours, in 7 of 13 answers.
Pastes a community garden club's web page and describes three problems on phones: the page scrolls sideways, section headings end up hidden under the sticky header after tapping the menu links, and three event cards are squashed instead of sitting one per row. It asks for the whole page back with all three fixed, leaving the laptop layout as it is and the header still sticky.
Every check passed: by 10 of 13 answers.
Missed most often:On a 375 px phone the page no longer scrolls sideways, in 3 of 13 answers.
What each model built
The top of each page, from our own screenshot at a desktop width, at each model’s highest setting. One answer has only a phone screenshot, shown narrow. Three answers ran out of room before writing a page. Open one for the answer, its checks and every screenshot.
Of the six models tested at both low and max effort, two scored higher here at max effort, four lower and no the same. The widest change was Sonnet 5.5, from 98 at low effort to 80 at max.
Score on building web pages at low and max effort, for each model with both
Each prompt asks for one finished web page in a single file: a landing page, a dashboard drawn from a picture of its design, a kanban board that keeps its cards after a reload, a Minesweeper game, and a fix for three layout bugs on a page that already exists.
Every page is opened in a real browser at a desktop width and a phone width. A program clicks through it, checks the console, the layout, the headings and anything loaded from outside the file, and takes the screenshots shown here. Where a check asks whether a page looks finished, a second model reads it from those screenshots.
Five prompts and 41 checks in all: 39 run or counted by a program and 2 read by a second model. The full method.
