MEGA-Bench Task - Transit Map Intersection Points: leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20240620) | 71.7 |
| 2 | Claude 3.5 Sonnet (20241022) | 66.1 |
| 3 | Gemini 2.5 Pro (Preview 03-25) | 61.6 |
| 4 | GPT-4o | 56.2 |
| 5 | Qwen 2 VL 72B | 51.9 |
| 6 | InternVL2.5-78B | 48.2 |
| 7 | GPT-4o Mini | 44 |
| 8 | Gemini 1.5 Flash (002) | 40.7 |
| 9 | Llama 4 Scout Base | 37.5 |
| 10 | Pixtral-12B | 35.7 |
| 11 | InternVL3-78B | 35.4 |
| 12 | InternVL3-38B | 34 |
| 13 | Gemma 3 27B (IT) | 33.9 |
| 14 | Gemini 1.5 Pro (002) | 33 |
| 15 | InternVL3-8B | 27.4 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-transit-map-intersection-points · How It Works · Data refreshed daily, snapshot 2026-09-05.