Benchmark
Design Arena Elo (models/video)
People compare model answers and choose which they prefer. Higher scores are better.
- Models measured
- 30
- Scores recorded
- 31
- Sources
- 1
- IQ weight
- 1.17
Score distribution
- 1,169.7
- Mean
- 1,158
- Median
- 98.7
- Std dev
- 1,060
- 10th percentile
- 1,281
- 90th percentile
- 937
- Min
This benchmark has a weight of 1.17 for video-generation models, measured over 30 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.