Dr Pia Model Bake-off

Cosmetic Dentistry NYC, content test v2, Oct 18, 2026

Round A: Writer

Same brief, same facts, same prompt. Only the writing model changes. 6 keywords, 3 runs per model.

Leaderboard

GLM 5.33 articles, 1 empty89.3spread 0.560.379.745.370.03.00.003/33/30$0.00
Grok 4.718 articles87.9spread 3.259.467.948.259.72.50.0018/1817/180$0.00
Kimi K318 articles84.3spread 3.559.276.452.973.03.20.0018/1818/180$0.00
Fable6 articles83.7spread 4.956.769.750.279.33.80.006/66/60$0.00
Claude Opus 5.518 articles81.8spread 3.861.675.851.277.83.70.0018/1818/181$0.00
Astra18 articles80.6spread 3.858.266.850.376.33.60.0018/1818/1821$0.00
Solhigh18 articles80.3spread 4.459.270.850.878.03.70.0018/1818/1817$0.00
DeepSeek V4 Pro (direct)6 articles80.0spread 3.160.574.052.873.03.20.006/65/62$0.00
GPT 6.1 Sol18 articles79.9spread 4.658.369.248.276.23.70.0018/1818/1814$0.00
Terra18 articles79.5spread 4.061.673.351.376.53.70.0018/1815/180$0.00
MiniMax M316 articles, 1 empty77.4spread 3.857.876.350.275.63.70.0016/166/160$0.00
Claude Sonnet 5.518 articles77.3spread 2.557.173.248.180.94.00.0018/1818/180$0.00
DeepSeek V4 Pro17 articles, 4 empty76.3spread 4.260.579.452.372.83.20.0017/177/1710$0.57

Click a column heading to sort. Pending means the score has not landed yet. It is never counted as zero. NeuronWriter, Surfer and Frase are shown side by side on purpose.

NeuronWriter, Surfer SEO, Frase and blind quality per model
Scores side by side