MEGA-Bench Task - Bridge Strategies Expert: leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20240620) | 41.4 |
| 2 | InternVL2-8B | 40 |
| 3 | Gemini 2.5 Pro (Preview 03-25) | 40 |
| 4 | Qwen 2 VL 2B | 36.4 |
| 5 | GPT-4o | 35.7 |
| 6 | Claude 3.5 Sonnet (20241022) | 35 |
| 7 | Aquila-VL-2B | 34.3 |
| 8 | Gemini 1.5 Pro (002) | 34.3 |
| 9 | GPT-4o Mini | 33.6 |
| 10 | InternVL3-78B | 32.9 |
| 11 | Gemini 1.5 Flash (002) | 32.1 |
| 12 | Llama 4 Scout Base | 30 |
| 13 | InternVL2.5-78B | 28.6 |
| 14 | InternVL3-38B | 27.9 |
| 15 | Gemma 3 27B (IT) | 24.3 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-bridge-strategies-expert · How It Works · Data refreshed daily, snapshot 2026-09-05.