claude-fable-5
claude-fable-5 (Anthropic) scored 81.9% 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 2026-07-09.
Agent harnessHigh reasoning
Composite
81.9%
95% CI 74.5% - 88.6%
Est. cost
$7.62
USD
Tokens
200,633
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.
Websites
15 tasks
45.8%
composite
svg-scene
1 task
65.6%
judge
65.6%
composite
Animation
2 tasks
72.7%
judge
72.7%
composite
Auth screen
1 task
73.4%
judge
73.4%
composite
Marketing page
1 task
82.8%
judge
82.8%
composite
SVG and graphics
13 tasks
86.6%
composite
Dashboard
1 task
89.1%
judge
89.1%
composite
UI components
19 tasks
89.2%
composite
Australian accounting
23 tasks
89.9%
composite
Australian law
5 tasks
100.0%
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
Objective tasks
Programming, Australian legal and accounting, graded by execution. 30 of 34 scored a perfect 100.0%; the rest are below. Open the answer in a popup, or compare it across every model.