VideoOdyssey-V - 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 visual subset questions about inferring causes and consequences; higher is better.
Metric: Multiple-choice accuracy (%) on the 128 causal reasoning questions (infer the underlying causes or resulting consequences of a specific event) of VideoOdyssey-V (human-written questions on 100 videos of 60 to 255 minutes that need visual cues and long continuous viewing; 1,618 in all), 128 frames for proprietary models and each open model's maximum frame setting; higher is better. Source: arxiv.org. Saturation forecast: Around July 2027. 18 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 59.4 | #102 |
| 2 | Seed 2.0 Pro | 57.8 | #226 |
| 3 | Kimi K2.5 | 57 | #272 |
| 4 | Gemini 2.5 Pro | 52.3 | #280 |
| 5 | GPT-5.2 | 51.6 | #190 |
| 6 | Qwen 3.5 27B | 48.4 | #380 |
| 7 | Qwen 3 VL 235B A22B | 45.3 | #486 |
| 8 | InternVL3.5-38B | 42.2 | #1026 |
| 9 | Video-KTR-7B | 41.4 | |
| 10 | Video-R1 | 37.5 | #1384 |
| 11 | Qwen 3 VL 32B | 34.4 | #624 |
| 12 | Phi-4 Multimodal Instruct | 32.8 | #4155 |
| 13 | VideoLLaMA3-7B | 30.5 | #2064 |
| 14 | Kimi-VL-A3B | 28.9 | #2362 |
| 15 | LLaVA-OneVision-1.5-8B | 27.3 | #1876 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=videoodyssey-v-causal-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-22.