Skip to main content
VerdictAI Benchmark
Back to Leaderboard
DeepSeek logo

DeepSeek R1 0528 Qwen3 8B

Open Weight

Provider: DeepSeek • ID: deepseek-r1-qwen3-8b

Detailed evaluation metrics and judge breakdown for DeepSeek R1 0528 Qwen3 8B across all 4 launch categories.

Compare in Blind Arena
Pricing: In — • Out —
Composite Verdict Score
10.4
0
.
0
/100

Category Score Breakdown

Frontend UI10.6

Interactive dashboards and landing pages

Game Dev10.2

Canvas 2D physics and browser games

SVG Art10.5

Vector graphics and generative math art

Agentic Tasks10.1

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
10.7
Craft— Clean semantic markup and clean TypeScript
10.5
Design— Visual aesthetics, contrast, layout harmony
10.3
Creativity— Original micro-interactions and layout depth
10.1
Fidelity— Strict adherence to complex prompt constraints
10.2