MEGA-Bench Task - Face Keypoint Detection: leaderboard
The "Face Keypoint Detection" 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: Not forecast. 44 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL2-8B | 84.9 | #817 |
| 2 | Aquila-VL-2B | 84.9 | #1238 |
| 3 | GLM-4.1V-9B (Thinking) | 80.1 | #472 (GLM-4.1V-9B) |
| 4 | Qwen 2 VL 7B | 75.4 | #750 |
| 5 | Gemini 1.5 Flash (002) | 74.6 | #583 |
| 6 | Llama 4 Scout Base | 69 | #707 |
| 7 | Claude 3.5 Sonnet (20241022) | 68.1 | #291 |
| 8 | Gemini 1.5 Pro (002) | 67.6 | #349 |
| 9 | Qwen 2 VL 72B | 67 | #460 |
| 10 | MiniCPM-V-2.6 | 65.1 | #836 |
| 11 | GPT-4o Mini | 65 | #599 |
| 12 | Pixtral-12B | 64.9 | #808 |
| 13 | InternVL3-38B | 63.3 | #405 |
| 14 | Gemini 2.5 Pro (Preview 03-25) | 61.7 | #123 |
| 15 | Gemma 3 27B (IT) | 60.4 | #523 |
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-face-keypoint-detection · How It Works · Data refreshed daily, snapshot 2026-10-20.