MEGA-Bench Task - Crossword Mini 5x5: leaderboard

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

Top models

#ModelScore
1Claude 3.5 Sonnet (20240620)77.1
2Gemini 2.5 Pro (Preview 03-25)75
3Gemini 1.5 Pro (002)73.6
4Claude 3.5 Sonnet (20241022)69.3
5GPT-4o67.9
6Gemma 3 27B (IT)52.1
7GPT-4o Mini48.6
8Qwen 2 VL 72B44.3
9Llama 4 Scout Base39.3
10InternVL3-78B35
11Gemini 1.5 Flash (002)35
12InternVL2.5-78B32.9
13Gemma 3 12B (IT)29.3
14InternVL3-38B22.1
15InternVL3-8B15.7

Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-crossword-mini-5x5 · How It Works · Data refreshed daily, snapshot 2026-09-05.