VideoOdyssey-AV - Causal Reasoning: 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 inferring causes and consequences; higher is better.

Metric: Multiple-choice accuracy (%) on the 58 causal reasoning questions (infer the underlying causes or resulting consequences of a specific event, based on 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 2027. 12 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Flash56.9#175
2Gemini 3.1 Pro (Preview)53.5#99
3Gemini 2.5 Pro50#276
4Qwen 3.5 Omni Plus50#254
5Ola-7B31#1557
6Qwen 3 Omni 30B A3B29.3#847
7VITA-1.5-7B24.1#1853
8Unified-IO-2 XXL22.4
9Unified-IO-2 L22.4
10Unified-IO-2 XL20.7
11VideoLLaMA2-7B17.2#2385
12OneLLM17.2#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-causal-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.