OmniStarPro-Long - LMR (Medium Span): leaderboard
Metric: Recall accuracy (%; share of long-range memory recall (LMR) queries, which ask late in the stream for an attribute of an entity or event seen early in it, that a GPT-4o judge marks as matching the ground truth; evidence 10-30 minutes before the query; OmniStarPro-Long partition, streams of 10 minutes to over an hour). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | LiveStarPro | 49.7 |
| 2 | LiveStar | 33 |
| 3 | MMDuet | 24.5 |
| 4 | VideoLLM-MoD | 19.3 |
| 5 | VideoLLM-online | 18.6 |
Interactive version: theaggregate.ai/benchmark?slug=omnistarpro-long-lmr-medium-span · How It Works · Data refreshed daily, snapshot 2026-09-26.