Best AI for Humor
Three minutes of stand-up, puns explicitly ruled out. Plus a satirical headline and a manifesto. Most models are not funny. Some are.
3 challenges20 models
How Humor rankings are computed
20 models tested across 3 humor challenges.Composite score: 30% Rival Index, 20% task coverage, 20% challenge-scoped duel performance, 15% recency, 15% tier. Deduplicated by product line. Gemini 3.1 Pro Preview leads at 82.5/100. Drawn from Rival's open dataset of 21,000+ human preference votes.
Rival's Pick#6 Rival IndexGoogle flagship
Too close to callComposite scores combine task evidence, Rival Index, recency, and model tier. Rival’s Pick is a separate editorial recommendation. How ranking works
GPT-6 Astra
openai
$10.00·$50.00
79Composite
Gemini 3.1 Pro Preview
google
$2.00·$12.00
82Composite
Gemini 3.8 Flash
google
$0.75·$3.75
77Composite
Head-to-Head
Full Rankings
20 models#
Model
Coverage
Index
Price
Composite
Challenges3
Related