VideoOdyssey-AV - Temporal Ordering: leaderboard
VideoOdyssey (2026) tests ultra-long video understanding with human-written multiple-choice questions on videos averaging 109 minutes, each annotated with how long a viewer must watch continuously to answer. This board is accuracy in percent on the audio-visual subset questions about ordering events in time; higher is better.
Metric: Multiple-choice accuracy (%) on the 50 temporal ordering questions (arrange multiple key events from the video according to both visual and audio cues) of VideoOdyssey-AV (human-written questions on 100 videos of 60 to 255 minutes that need both visual and audio cues; 1,062 in all), omni-modal models given video and audio, 64 frames (VITA-1.5-7B 16); higher is better. Source: arxiv.org. Saturation forecast: Around February 2028. 12 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 44 | #175 |
| 2 | Gemini 2.5 Pro | 38 | #276 |
| 3 | Gemini 3.1 Pro (Preview) | 34 | #99 |
| 4 | Unified-IO-2 L | 32 | |
| 5 | VideoLLaMA2-7B | 26 | #2385 |
| 6 | Unified-IO-2 XL | 24 | |
| 7 | Qwen 3.5 Omni Plus | 20 | #254 |
| 8 | Ola-7B | 20 | #1557 |
| 9 | Unified-IO-2 XXL | 20 | |
| 10 | VITA-1.5-7B | 20 | #1853 |
| 11 | Qwen 3 Omni 30B A3B | 18 | #847 |
| 12 | OneLLM | 16 | #2372 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=videoodyssey-av-temporal-ordering · How It Works · Data refreshed daily, snapshot 2026-11-21.