HAVEN - Saliency Ranking: leaderboard

Saliency ranking on HAVEN (2026), a video understanding benchmark with aligned frames, shots, descriptions and full video documents: the model ranks video shots by importance from their frames alone. The score is the Spearman rank correlation with the annotated ranking, from -1 to 1; higher is better.

Metric: Shot saliency ranking: Spearman rank correlation (-1 to 1) between the predicted and the true order by importance, frames only, zero-shot on a random sample of 100 videos per source dataset; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 6 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-5.20.21#185
2Qwen 3 VL 8B Instruct0.187#1029
3InternVideo2.5-Chat-8B0.169#2036
4GLM-4.6V-Flash0.164#1040
5VideoLLaMA3-7B0.142#2050
6Qwen 2.5 VL 7B Instruct0.006#1902

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=haven-saliency-ranking · How It Works · Data refreshed daily, snapshot 2026-11-21.