Independent AI marketing benchmark
See which AI does the best marketing work.
Compare anonymous model responses to real marketing briefs. Your vote improves a public, provider-neutral ranking.
- Models
- —
- Model evaluations
- —
- Task types
- 6
Each participating model counts once per ranked vote. Evaluations do not count unique voters.
Blind comparison
Which response is stronger?
Loading matchup...
Beyond the leaderboard
Research tools for evaluating AI marketing.
Run a controlled evaluation or inspect where leading models agree. Each tool keeps the method visible.
Evaluate
Custom evaluations
Score anonymous model work against a weighted rubric you define.
Build an evaluationResearch
Tool recommendations
Compare provider-balanced recommendations for real marketing stack decisions.
Ask the model panelLibrary
Prompt trends
Inspect the reusable prompts models recommend across marketing roles.
Browse prompt trendsLive data
Current Standings
- 01
GPT-5
OpenAI · provisional
1262 Elo
100.0% · 4 votes
- 02
GPT-5.2
OpenAI · provisional
1231 Elo
100.0% · 2 votes
- 03
Claude Sonnet 4.6
Anthropic · provisional
1229 Elo
75.0% · 4 votes
- 04
Gemini 3.1 Flash Lite Preview
Google · provisional
1227 Elo
75.0% · 4 votes
- 05
GPT-5.1
OpenAI · provisional
1215 Elo
100.0% · 1 vote
Elo ratings from blind pairwise votes. How it works