OmniStarPro-RNG (Online) - Summary Fluency: leaderboard
Narrative quality of self-timed real-time narration on OmniStarPro (Yang et al., 2026): a GPT-4o judge compares the concatenated narration with the ground-truth narrative for logic, fluency, conciseness, consistency and completeness, on a 0-10 scale. Higher is better.
Metric: Summary fluency (0-10; GPT-4o judge rating of the concatenated self-timed narration against the ground-truth narrative on logicality, fluency, conciseness, semantic consistency and completeness). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | LiveStarPro | 4.41 | |
| 2 | LiveStar | 4.25 | #1076 |
| 3 | MMDuet | 2.69 | |
| 4 | VideoLLM-online | 0.59 | |
| 5 | VideoLLM-MoD | 0.55 |
Interactive version: theaggregate.ai/benchmark?slug=omnistarpro-rng-online-summary-fluency · How It Works · Data refreshed daily, snapshot 2026-11-12.