4,700-Vote Blind Marketing Eval: GPT-6.2 Sol Leads Overall but All Three Top Models Statistically Tie
boringmarketer · x · 2026-11-21
Boring Marketing ran the deepest AI marketing eval yet: 35 real brands, 110 real assets, 4,700+ blind judge votes from 5 labs with no judge scoring its own lab's model. Key findings:
- Statistically tied overall: GPT-6.2 Sol (1022), Claude Opus 5.5 (999), Gemini 4 Neon (979) — 95% ranges overlap.
- By channel: Sol wins social posts (67% on Instagram, 65% on X) and total votes (58% avg across 15 jobs vs Opus 52%, Gemini 40%) at the lowest cost ($0.04/piece); Opus leads video editing (77%) and Meta ads (64%); Gemini leads video ads (75%) and YouTube packaging (61%).
- Deep work: Sol and Opus tied at top, Gemini trails; judges favor Sol (57%) on copywriting while outside judges favor Opus (66%).
- Most surprising: the brands' own best work still beat all three models most of the time.
More from Models
- GPT-6.2 Astra review: stronger at math and economic modeling, now with Ultrafast — soumitrashukla9 · 2026-11-21
- Persimmon v0.1 Pro/Ultra launch: 8-turn Turing test score doubles from 18% to 48% — gharik · 2026-11-21
- Google launches first interactive Gemma 4 explainer teaching how LLMs work in the browser — DynamicWebPaige · 2026-11-21
- InferenceBench recap: Anthropic led for half a year before Google's dramatic comeback — maksym_andr · 2026-11-21
- Persimmon built a safety framework from scratch for human-like user models — gharik · 2026-11-21
- Persimmon is now open to everyone, safety filters credited for the decision — gharik · 2026-11-21