ProactiveVideoQA AURA Study - VAD Omega 1: leaderboard
The AURA 2.0 paper compares released models on streaming video interaction. IPIBench, StreamArena, ProactiveVideoQA and PhoStream retain their own protocols. ProactiveVideoQA reports PAUC; PhoStream’s overall score is question-weighted, and its timing categories are not guarantees of answer correctness. This board reports pAUC (native). Higher is better.
Metric: PAUC (native). Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 Mini | 54 | #781 |
| 2 | GPT-4.1 | 51.6 | #500 |
| 3 | MMDuet (Harness: remove assistant turns) | 46.5 | |
| 4 | Gemini 2.0 Flash | 38.6 | #717 |
| 5 | Gemini 1.5 Pro | 37.9 | #1061 |
| 6 | JoyAI-VL-Interaction | 32.9 | |
| 7 | LongVA-7B | 32.3 | #2788 |
| 8 | Qwen 2.5 VL 7B | 30.7 | #1838 |
| 9 | MMDuet | 28.5 | |
| 10 | InternLM-XComposer2.5 | 28.3 | #3981 |
| 11 | LLaVA-OneVision-7B | 25.9 | #2685 |
| 12 | VideoLLM-online-8B | 25 | |
| 13 | InternVL2.5-8B | 20.8 | #2444 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=proactivevideoqa-aura-study-vad-omega-1 · How It Works · Data refreshed daily, snapshot 2026-11-16.