MEGA-Bench Task - Visual Prediction Rater Plane Segmentation — leaderboard

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

Top models

#ModelScore
1Gemini 2.5 Pro95.6
2Claude 3.5 Sonnet (20240620)62.2
3Gemma 3 27B (IT)55.6
4Gemini 2.0 Flash (Preview)51.1
5Claude 3.5 Sonnet (20241022)48.9
6InternVL3-38B44.4
7Gemini 1.5 Pro (002)40
8GPT-4o31.1
9Gemini 1.5 Flash (002)28.9
10InternVL3-78B26.7
11Llama 4 Scout Base24.4
12Qwen 2 VL 72B22.2
13Aquila-VL-2B22.2
14InternVL3-14B20
15Pixtral-12B17.8

Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-visual-prediction-rater-plane-segmentation · How the rankings work · Data refreshed daily, snapshot 2026-07-22.