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 Pro75
3Gemini 1.5 Pro (002)73.6
4Claude 3.5 Sonnet (20241022)69.3
5GPT-4o67.9
6Gemini 2.0 Flash (Preview)67.1
7Gemma 3 27B (IT)52.1
8GPT-4o Mini48.6
9Qwen 2 VL 72B44.3
10Llama 4 Scout Base39.3
11InternVL3-78B35
12Gemini 1.5 Flash (002)35
13InternVL2.5-78B32.9
14InternVL3-14B30.7
15Gemma 3 12B (IT)29.3

Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-crossword-mini-5x5 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.