Artificial Intelligence1 min read

Two stealth models just appeared on the leaderboard, both ahead of Fable 5.2

By · Published by Everything Blog

In short

The Frontier Design Benchmark tests creative capabilities using a comprehensive framework powered by blind comparisons and detailed evaluations. To date, 68 professional designers have contributed 4,311 votes across 33 unique marketplace briefs. The platform pairs Elo ratings with a 35-dimension design critique rubric that holds a 0.94 correlation with designer ratings. This testing methodology also tracks artificial intelligence slop patterns to provide explainable assessments of model outputs.

Key points

  • Professional human designers evaluate frontier artificial intelligence models by comparin…: Professional human designers evaluate frontier artificial intelligence models by comparing blind outputs directly against golden human designs.
  • Pairwise preference voting generates Elo ratings to assess design quality against models…: Pairwise preference voting generates Elo ratings to assess design quality against models like Claude Opus 5.5 and GPT-6.1 Sol.
  • Contra Labs integrates an artificial intelligence slop pattern list to deliver explainabl…: Contra Labs integrates an artificial intelligence slop pattern list to deliver explainable assessments of creative capabilities.

Sources checked by our analysis pipeline: 1

Deconstructing Real World Creativity.

Exploring the frontier of AI design capabilities through real client briefs from Contra’s marketplace, golden designs and model comparisons.

Leaderboard

How frontier models rank in blind head-to-head votes on real client briefs from Contra’s marketplace.

Frontier

Design Benchmark

Elo rating from blind head-to-head votes

- Gestalt

- Human golden

- Claude Fable 5.2

- Claude Opus 5.5

- GPT-6.1 Sol

- Claude Sonnet 5.5

- Kimi K3.1 Pro

- GPT-6 Astra

- Qwen3.8 Max

- Kimi K3

- GLM-5.3 Prime

- Gemini 4 Carbon

- Muse Spark 1.3

- Muse Wonder

- Grok 4.7

- Gemini 4 Argon

- DeepSeek V4 Pro

contralabs.com/frontierdesign

Real design projects, judged by designers and scored by rubrics

Frontier Design Benchmark is the first benchmark built on real world, high value, and complex design projects. It is judged by professional designers in the Contra Labs network, ensuring the quality of evaluation.

In the benchmark, we also built a design critique rubric and a list of AI slop patterns. The rubric scores are highly correlated with the designers’ ratings of the models…

Artificial Intelligence