MEGA-Bench Task - Remaining Playback Time Calculation — leaderboard

Metric: Task Score (%). Source: huggingface.co. 44 models tracked.

Top models

#ModelScore
1Ovis2-8B35.7
2GPT-4o28.6
3InternVL3-78B28.6
4GPT-4o Mini28.6
5InternVL2.5-78B28.6
6Llama 4 Scout Base28.6
7InternVL3-14B21.4
8Claude 3.5 Sonnet (20240620)21.4
9Claude 3.5 Sonnet (20241022)14.3
10Gemini 2.5 Pro7.1
11Gemma 3 27B (IT)7.1
12InternVL3-38B7.1
13Pixtral-12B7.1
14Gemini 1.5 Flash (002)7.1
15Gemma 3 12B (IT)0

Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-remaining-playback-time-calculation · How the rankings work · Data refreshed daily, snapshot 2026-07-22.