MEGA-Bench Task - Distinguish AI Generated Image — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 94.7 |
| 2 | Ovis2-8B | 84.2 |
| 3 | Gemini 1.5 Flash (002) | 84.2 |
| 4 | Gemini 2.0 Flash (Preview) | 84.2 |
| 5 | Gemma 3 27B (IT) | 78.9 |
| 6 | Qwen 2 VL 7B | 78.9 |
| 7 | InternVL2.5-78B | 78.9 |
| 8 | InternVL3-78B | 73.7 |
| 9 | GPT-4o | 68.4 |
| 10 | InternVL3-8B | 68.4 |
| 11 | Qwen 2 VL 72B | 68.4 |
| 12 | InternVL3-38B | 68.4 |
| 13 | MiniCPM-V-2.6 | 68.4 |
| 14 | Gemini 1.5 Pro (002) | 68.4 |
| 15 | Llama 4 Scout Base | 68.4 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-distinguish-ai-generated-image · How the rankings work · Data refreshed daily, snapshot 2026-07-22.