MEGA-Bench Task - ILI Ratio Future Prediction — leaderboard

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

Top models

#ModelScore
1Claude 3.5 Sonnet (20240620)34
2Gemini 1.5 Pro (002)26.9
3Claude 3.5 Sonnet (20241022)23.9
4InternVL3-38B18.1
5Gemini 2.0 Flash (Preview)16.1
6GPT-4o12.5
7Gemini 2.5 Pro11.6
8Gemini 1.5 Flash (002)11.6
9InternVL3-8B10.2
10Llama 4 Scout Base9
11InternVL3-78B6.3
12Gemma 3 4B (IT)5.6
13Qwen 2 VL 72B5.3
14Gemma 3 27B (IT)3.9
15InternVL2.5-78B2.2

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