Claude 4 Sonnet (Non-reasoning)
Provider: Anthropic • ID: claude-4-sonnet
Detailed evaluation metrics and judge breakdown for Claude 4 Sonnet (Non-reasoning) across all 4 launch categories.
Compare in Blind Arena
Pricing: In $3.00 per 1M tokens • Out $15.00 per 1M tokens
Composite Verdict Score/100
25.5
0
.0
Category Score Breakdown
Frontend UI26.0
Interactive dashboards and landing pages
Game Dev25.0
Canvas 2D physics and browser games
SVG Art25.7
Vector graphics and generative math art
Agentic Tasks24.7
Multi-step execution planning
Auditable Multi-Judge Dimension Scoring
Grades evaluated across 3 independent judge models using published rubrics.
Functionality— Executes without runtime JS exceptions
26.3
Craft— Clean semantic markup and clean TypeScript
25.7
Design— Visual aesthetics, contrast, layout harmony
25.2
Creativity— Original micro-interactions and layout depth
24.7
Fidelity— Strict adherence to complex prompt constraints
25.0