Frontier Design Benchmark: Claude Fable 5.2 tops ranking, just 26 Elo below human designs
yuwen_lu_ · x · 2026-11-21
Contra Labs launched the Frontier Design Benchmark, arguing that model design capability should be judged by professional designers voting blind on real client briefs rather than random internet users.
- Of 15 models tested, Claude models took 3 of the top 4: #1 Claude Fable 5.2 (1713), #2 Claude Opus 5.5 (1691), #4 Claude Sonnet 5.5 (1667)
- Claude Fable 5.2 sits within 26 Elo of the human golden baseline (1739)
- Others: GPT-6.1 Sol (1668), Kimi K3.1 Pro (1663), GPT-6 Astra (1638), Qwen3.8 Max (1434), DeepSeek V4 Pro (1121) at the bottom
Note: several model names on the board (e.g. Claude Fable 5.2, GPT-6.1) are unverified against known releases.
More from Models
- Kaomoji emotion probe validated: 0.68 correlation with valence direction, 0.79 when trained — Sauers_ · 2026-11-21
- Kaomoji as an emotion probe: Qwen3-8B's internal "happiness" is measurable — Sauers_ · 2026-11-21
- 27,262 thinking tokens, zero answer: Google auto-billed 23 empty Vertex AI crashes via HTTP 200 — Plenty-Jackfruit-218 · 2026-11-21
- Thinking Machines' new model claimed 5x faster than Qwen3.8, 15% more accurate — LiTianleli · 2026-11-21
- Gemini silently saves every Google search that gets an AI response, user finds — The-Neofighter · 2026-11-21
- Hy4 Flash reads 1M tokens in 44 min on Mac Studio M3 Ultra, 81 min on DGX Spark — AIFlow_ML · 2026-11-21