VideoOdyssey-AV - Summarization: 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 summarizing long stretches; higher is better.
Metric: Multiple-choice accuracy (%) on the 50 summarization questions (analyze the audio-visual information to achieve a high-level, abstract understanding) 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 2027. 12 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 64 | #99 |
| 2 | Qwen 3.5 Omni Plus | 62 | #254 |
| 3 | Gemini 2.5 Pro | 60 | #276 |
| 4 | Gemini 3 Flash | 50 | #175 |
| 5 | Ola-7B | 36 | #1557 |
| 6 | Qwen 3 Omni 30B A3B | 34 | #847 |
| 7 | VideoLLaMA2-7B | 32 | #2385 |
| 8 | Unified-IO-2 XL | 28 | |
| 9 | OneLLM | 24 | #2372 |
| 10 | Unified-IO-2 L | 24 | |
| 11 | Unified-IO-2 XXL | 20 | |
| 12 | VITA-1.5-7B | 14 | #1853 |
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-summarization · How It Works · Data refreshed daily, snapshot 2026-11-21.