VideoOdyssey-V - Attribute 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 visual attributes; higher is better.

Metric: Multiple-choice accuracy (%) on the 106 attribute recognition questions (identify the specific visual attributes of entities, including humans, objects and scenes) 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 February 2027. 18 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3.1 Pro (Preview)64.2#102
2Seed 2.0 Pro61.3#226
3Qwen 3.5 27B59.4#380
4Gemini 2.5 Pro54.7#280
5GPT-5.251.9#190
6Kimi K2.551.9#272
7Qwen 3 VL 235B A22B49.1#486
8Qwen 3 VL 32B49.1#624
9InternVL3.5-38B38.7#1026
10VideoLLaMA3-7B37.7#2064
11Video-R134#1384
12LLaVA-NeXT-Video-DPO (34B)34#2493
13LLaVA-OneVision-1.5-8B33#1876
14Video-KTR-7B33
15Phi-4 Multimodal Instruct32.1#4155

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-attribute-recognition · How It Works · Data refreshed daily, snapshot 2026-11-22.