MEGA-Bench Task - Booking Web Recommendation — leaderboard

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

Top models

#ModelScore
1GPT-4o78
2Claude 3.5 Sonnet (20240620)76.2
3Gemini 2.5 Pro74.2
4Gemini 2.0 Flash (Preview)71.2
5GPT-4o Mini70.6
6Gemini 1.5 Pro (002)67.7
7Claude 3.5 Sonnet (20241022)67.5
8InternVL2.5-78B63.1
9Gemini 1.5 Flash (002)62
10Pixtral-12B59.9
11Gemma 3 12B (IT)55.3
12InternVL3-14B55
13Gemma 3 27B (IT)54
14Llama 4 Scout Base48.3
15Gemma 3 4B (IT)48.1

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