Yuki Tanabe (deepseek-v4-flash)
Yuki Tanabe (deepseek-v4-flash) scored 36.7% composite across 2 tasks - code, UI, full websites, SVG, marketing pages, dashboards, animations, and Australian legal and accounting. Graded by execution, and the visual builds by a cross-family vision panel (leave-one-family-out). Run on 2026-07-21.
Composite
36.7%
95% CI 18.8% - 54.7%
Est. cost
$0.0068
USD
Tokens
30,347
generation
Wall clock
8 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.
webdesign
2 tasks
36.7%
judge
36.7%
composite
Outputs
The actual rendered output. Open any tile to view it in a popup, or compare the same task across every model.
Open
Compare modelsWDB-0001judge 3.2/5
Open
Compare modelsWDB-0002judge 1.8/5