Video Understanding on PerceptionTest (test)
82.7AccuracyPLM-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| PLM-8BCategory=Fully open2026.03 | 82.7 | |
| Molmo2-8BCategory=Fully open2026.03 | 82.1 | |
| Molmo2-4BCategory=Fully open2026.03 | 81.3 | |
| Eagle2.5-8BCategory=Open weights2026.03 | 81 | |
| Molmo2-O-7BCategory=Fully open2026.03 | 79.6 | |
| GPT-5Category=API call only2026.03 | 79.4 | |
| PLM-3BCategory=Fully open2026.03 | 79.3 | |
| Gemini 2.5 ProCategory=API call only2026.03 | 78.4 | |
| Gemini 3 ProCategory=API call only2026.03 | 77.6 | |
| VideoChat-Flash-7BCategory=Fully open2026.03 | 76.5 | |
| Gemini 2.5 FlashCategory=API call only2026.03 | 74.7 | |
| GLM-4.1V-9BCategory=Open weights2026.03 | 74.2 | |
| InternVL3.5-8BCategory=Open weights2026.03 | 72.7 | |
| Qwen3-VL-8BCategory=Open weights2026.03 | 72.7 | |
| GPT-5 miniCategory=API call only2026.03 | 72 | |
| MiniCPM-V-4.5-8BCategory=Open weights2026.03 | 70.9 | |
| Qwen3-VL-4BCategory=Open weights2026.03 | 70.7 | |
| LLaVA-Video-7BCategory=Fully open2026.03 | 68.8 | |
| InternVL3.5-4BCategory=Open weights2026.03 | 68.1 | |
| Claude Sonnet 4.5Category=API call only2026.03 | 64.3 | |
| Keye-VL-1.5-8BCategory=Open weights2026.03 | 64.2 |