MEGA-Bench Task - Pictionary Skribbl.io — leaderboard

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

Top models

#ModelScore
1Gemini 2.5 Pro60
2GPT-4o55
3Gemini 1.5 Pro (002)35
4Gemini 2.0 Flash (Preview)30
5Claude 3.5 Sonnet (20241022)25
6GPT-4o Mini20
7Qwen 2 VL 72B15
8Gemini 1.5 Flash (002)15
9InternVL2.5-78B15
10Claude 3.5 Sonnet (20240620)15
11Llama 4 Scout Base15
12InternVL3-78B10
13Ovis2-8B10
14Gemma 3 27B (IT)5
15Qwen 2 VL 7B5

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