MEGA-Bench Task - GUI Chat Hard — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Ovis2-8B | 70.4 |
| 2 | Qwen 2 VL 72B | 61 |
| 3 | InternVL2-8B | 60.3 |
| 4 | InternVL3-8B | 60 |
| 5 | Gemini 2.5 Pro | 58.8 |
| 6 | InternVL3-38B | 56.5 |
| 7 | InternVL3-78B | 54.2 |
| 8 | InternVL3-14B | 53.1 |
| 9 | Gemma 3 27B (IT) | 51.2 |
| 10 | Llama 4 Scout Base | 50.4 |
| 11 | GPT-4o | 49.4 |
| 12 | MiniCPM-V-2.6 | 48.4 |
| 13 | InternVL2.5-2B | 48.1 |
| 14 | Claude 3.5 Sonnet (20241022) | 45.8 |
| 15 | InternVL2.5-78B | 45.8 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-gui-chat-hard · How the rankings work · Data refreshed daily, snapshot 2026-07-22.