CapRiCorn-1K-V - Referential Consistency: leaderboard
Metric: Subject referential consistency (%; share of same-subject description pairs a GPT-4.1 judge clusters as one referent, over all pairs of that subject's vision-only keypoints, 1,000 videos). Source: arxiv.org. Saturation forecast: Around 2031. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 8B | 5.1 |
| 2 | Qwen 3.5 9B | 3.1 |
| 3 | Qwen 3.6 27B | 3.1 |
| 4 | Qwen 3.6 35B A3B | 2.9 |
| 5 | Qwen 3.5 122B A10B | 2.6 |
Interactive version: theaggregate.ai/benchmark?slug=capricorn-1k-v-referential-consistency · How It Works · Data refreshed daily, snapshot 2026-09-26.