MEGA-Bench Task - Media QA Web Stack Overflow: leaderboard
The "Media QA Web Stack Overflow" 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. 27 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2 VL 7B | 69 | #750 |
| 2 | GPT-4o Mini | 66.7 | #599 |
| 3 | Claude 3.5 Sonnet (20240620) | 66.7 | #299 |
| 4 | Qwen 2 VL 72B | 66.7 | #460 |
| 5 | InternVL2.5-78B | 66.7 | #431 |
| 6 | InternVL2-8B | 64.3 | #817 |
| 7 | GPT-4o | 61.9 | #343 |
| 8 | Claude 3.5 Sonnet (20241022) | 59.5 | #291 |
| 9 | Pixtral-12B | 59.5 | #808 |
| 10 | Gemini 1.5 Pro (002) | 57.1 | #349 |
| 11 | Gemini 1.5 Flash (002) | 54.8 | #583 |
| 12 | MiniCPM-V-2.6 | 42.9 | #836 |
| 13 | Aquila-VL-2B | 35.7 | #1238 |
| 14 | Qwen 2 VL 2B | 33.3 | #1220 |
| 15 | Llama 3.2 11B | 33.3 | #1199 |
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-media-qa-web-stack-overflow · How It Works · Data refreshed daily, snapshot 2026-10-20.