MEGA-Bench Task - Background Change: leaderboard
The "Background Change" subtask of MEGA-Bench Task. A single task from MEGA-Bench, a large-scale multimodal evaluation suite that scores vision-language models across 500+ real-world tasks spanning perception, reasoning, OCR, charts, documents, and multi-image understanding.
Metric: Task Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 03-25) | 92.9 | #123 |
| 2 | Gemma 3 27B (IT) | 85.7 | #523 |
| 3 | InternVL3-38B | 85.7 | #405 |
| 4 | Gemini 2.0 Flash (Preview) | 85.7 | #370 |
| 5 | Gemma 3 12B (IT) | 78.6 | #664 |
| 6 | InternVL3-8B | 78.6 | #620 |
| 7 | InternVL3-78B | 78.6 | #352 |
| 8 | Gemma 3 4B (IT) | 71.4 | #992 |
| 9 | GLM-4.1V-9B (Thinking) | 71.4 | #472 (GLM-4.1V-9B) |
| 10 | InternVL3-14B | 64.3 | #505 |
| 11 | Llama 4 Scout Base | 57.1 | #707 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-background-change · How It Works · Data refreshed daily, snapshot 2026-10-20.