GPT-6.2 Astra tops Design Creativity Bench at 70.4, Sol close at 4.5x lower cost
sandersted · x · 2026-11-19
Kombai's Design Creativity Bench (80 open-ended UI tasks × 2 product domains, 160 briefs with expert human designers as reference) has a new SOTA: GPT-6.2 Astra leads with a creativity score of 70.4, followed by GPT-6.2 Sol at 70.0.
The bigger story is Sol's catch-up: GPT-6.1 Sol trailed GPT-6 Astra by 1.8 points; with 6.2 the gap narrows to 0.4 points while Sol costs 4.5x less per design—both models now sit on the creativity-vs-cost Pareto frontier. The lead over Anthropic also grew: GPT-6.2 Sol scores nearly 20 points above Claude Opus 5.5 (vs a 13-point lead for 6.1 Sol), at 1/4 the cost of Opus and 1/6 that of Fable 5.2.
All models still lag human designers (92.3 reference; model originality 0.592 vs human 0.764). Grok 4.7 (62.2) and Gemini 3.8 Flash (62.0) round out the top ranks.
More from Models
- Nat Lambert rips White House plan to disclose 'Chinese model' use: where do you draw the line? — ivan_bezdomny · 2026-11-19
- Polymarket gives Google 79% odds of leading AI by year-end, OpenAI just 2.8% — Polymarket · 2026-11-19
- Heavy user pushes back on Astra 6.2 hate: big jump over 6.0, uses only 30% of quota — soumitrashukla9 · 2026-11-19
- Google buys bankrupt Spirit Airlines' internal workplace data for $10M to train AI models — Polymarket · 2026-11-19
- Google's Logan Kilpatrick demos Gemini 4: games, Android apps and simulations from one prompt — Shruti_0810 · 2026-11-19
- REAX 1.3 decision model, built with SN92 miners, listed on EU gateway Opper at $0.40/M — airesearch12 · 2026-11-19