MEGA-Bench Task - Visual Prediction Rater Depth Estimation — leaderboard

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

Top models

#ModelScore
1Gemini 2.5 Pro78.6
2GPT-4o73.8
3Gemini 1.5 Pro (002)66.7
4Gemini 2.0 Flash (Preview)61.9
5Claude 3.5 Sonnet (20241022)59.5
6Claude 3.5 Sonnet (20240620)47.6
7InternVL3-78B45.2
8Gemma 3 27B (IT)42.9
9InternVL3-38B38.1
10Llama 4 Scout Base32.1
11InternVL2.5-78B31
12Qwen 2 VL 72B28.6
13Aquila-VL-2B28.6
14Gemma 3 4B (IT)23.8
15MiniCPM-V-2.621.4

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