gpt-5.4-mini
gpt-5.4-mini (OpenAI) scored 76.0% composite across 87 tasks. This run covers Australian accounting, Animation, Auth screens, Dashboards, Australian law, Marketing pages, Programming, SVG and graphics, svg-scene, UI components, Websites. Published 2026-06-23.
API single-shotHigh reasoning
Composite
76.0%
95% CI 68.3% - 83.6%
Est. cost
$4.0799
USD
Tokens
936,673
generation
Wall clock
6 min
end to end
Gate pass
-
validated
Separation
-
good vs bad
By domain
Composite score per domain, weakest first. Judge is the vision model’s read, shown for the visual domains.
svg-scene
1 task
52.1%
judge
52.1%
composite
Websites
15 tasks
55.2%
composite
Australian accounting
23 tasks
59.4%
composite
Animation
2 tasks
61.5%
judge
61.5%
composite
Dashboard
1 task
70.8%
judge
70.8%
composite
Marketing page
1 task
75.0%
judge
75.0%
composite
Auth screen
1 task
79.2%
judge
79.2%
composite
Australian law
5 tasks
80.0%
composite
UI components
19 tasks
94.1%
composite
SVG and graphics
13 tasks
94.2%
composite
Programming
6 tasks
100.0%
composite
Outputs
The actual rendered output. Open any tile to view it in a popup, or compare the same task across every model.
53 results
![gpt-5.4-mini SVG and graphics output - Bar chart with exact proportional heights for [4, 10, 6, 8, 12]](/lab-thumbs/gpt-5-4-mini/DIAGR-0001.jpg)
![gpt-5.4-mini SVG and graphics output - Tiny bar chart for values [3, 7, 5, 9]](/lab-thumbs/gpt-5-4-mini/SVG-0005.jpg)
Objective tasks
Programming, Australian legal and accounting, graded by execution. 20 of 34 scored a perfect 100.0%; the rest are below. Open the answer in a popup, or compare it across every model.