MEGA-Bench Task - Hotel Booking Confirmation Parsing — leaderboard

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

Top models

#ModelScore
1Claude 3.5 Sonnet (20241022)76.4
2Gemini 2.5 Pro72.9
3GPT-4o71.4
4Claude 3.5 Sonnet (20240620)70.7
5Gemini 2.0 Flash (Preview)70.7
6InternVL2.5-78B66.4
7Gemma 3 27B (IT)65
8Qwen 2 VL 72B65
9Llama 4 Scout Base65
10InternVL3-8B63.6
11Ovis2-8B63.6
12GPT-4o Mini63.6
13Qwen 2 VL 7B62.9
14InternVL3-14B57.9
15InternVL3-78B57.1

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