MEGA-Bench Task - CLEVRER Moving Direction Video: leaderboard

The "CLEVRER Moving Direction Video" subtask of MEGA-Bench Task. A single task from MEGA-Bench, a large-scale multimodal evaluation suite that scores vision-language models across 500+ real-world tasks spanning perception, reasoning, OCR, charts, documents, and multi-image understanding.

Metric: Task Score (%). Source: huggingface.co. Saturation forecast: Around January 2027. 44 models tracked.

Top models

#ModelScoreOverall rank
1InternVL3-8B56.2#620
2InternVL2.5-78B56.2#431
3InternVL3-14B43.8#505
4Claude 3.5 Sonnet (20241022)37.5#291
5Claude 3.5 Sonnet (20240620)37.5#299
6InternVL3-38B37.5#405
7InternVL3-78B37.5#352
8Gemini 2.5 Pro (Preview 03-25)37.5#123
9GPT-4o Mini31.2#599
10Qwen 2 VL 72B31.2#460
11Llama 4 Scout Base31.2#707
12GPT-4o25#343
13Gemini 1.5 Pro (002)18.8#349
14GLM-4.1V-9B (Thinking)18.8#472 (GLM-4.1V-9B)
15Gemini 2.0 Flash (Preview)18.8#370

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-clevrer-moving-direction-video · How It Works · Data refreshed daily, snapshot 2026-10-20.