MEGA-Bench Task - Multilingual News QA: leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 1.5 Pro (002) | 100 |
| 2 | GPT-4o | 92.9 |
| 3 | Gemini 2.5 Pro (Preview 03-25) | 92.9 |
| 4 | Gemma 3 27B (IT) | 85.7 |
| 5 | Claude 3.5 Sonnet (20240620) | 85.7 |
| 6 | Gemma 3 12B (IT) | 78.6 |
| 7 | Claude 3.5 Sonnet (20241022) | 78.6 |
| 8 | Pixtral-12B | 71.4 |
| 9 | InternVL3-78B | 64.3 |
| 10 | Qwen 2 VL 72B | 64.3 |
| 11 | Gemini 1.5 Flash (002) | 64.3 |
| 12 | InternVL3-38B | 57.1 |
| 13 | GPT-4o Mini | 57.1 |
| 14 | InternVL2.5-78B | 50 |
| 15 | InternVL3-8B | 42.9 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-multilingual-news-qa · How It Works · Data refreshed daily, snapshot 2026-09-05.