VideoOdyssey-V - Action Recognition: 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 recognizing actions; higher is better.
Metric: Multiple-choice accuracy (%) on the 192 action recognition questions (identify the actions of humans or other objects in the video) 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 March 2027. 18 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 57.8 | #99 |
| 2 | Seed 2.0 Pro | 54.7 | #222 |
| 3 | Kimi K2.5 | 50 | #266 |
| 4 | Qwen 3.5 27B | 50 | #375 |
| 5 | Gemini 2.5 Pro | 49.5 | #276 |
| 6 | GPT-5.2 | 46.9 | #185 |
| 7 | Qwen 3 VL 235B A22B | 44.3 | #485 |
| 8 | InternVL3.5-38B | 38 | #1021 |
| 9 | Qwen 3 VL 32B | 34.9 | #622 |
| 10 | Video-KTR-7B | 33.9 | |
| 11 | Video-R1 | 33.3 | #1379 |
| 12 | VideoLLaMA3-7B | 29.2 | #2050 |
| 13 | LLaVA-NeXT-Video-DPO (34B) | 26.6 | #2489 |
| 14 | LLaVA-OneVision-1.5-8B | 25 | #1864 |
| 15 | Phi-4 Multimodal Instruct | 24.5 | #4154 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=videoodyssey-v-action-recognition · How It Works · Data refreshed daily, snapshot 2026-11-21.