Skip to content

Benchmark

Design Arena Elo (models/videotovideo)

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
8
Scores recorded
8
Sources
1
IQ weight
1.11

Score distribution

1,149.4
Mean
1,226
Median
155.8
Std dev
991
10th percentile
1,327
90th percentile
903
Min
9031,341
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.11 for video-generation models, measured over 7 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard