Video Captioning on Molmo2 CapTest (test)
56.6F1 ScoreGPT-5 mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5 miniAccess Type=API call only2026.01 | 56.6 | |
| GPT-5 miniCategory=API call only2026.03 | 56.6 | |
| GPT-5Access Type=API call only2026.01 | 50.1 | |
| GPT-5Category=API call only2026.03 | 50.1 | |
| Gemini 2.5 FlashAccess Type=API call only2026.01 | 46 | |
| Gemini 2.5 FlashCategory=API call only2026.03 | 46 | |
| Molmo2-8BAccess Type=Molmo2 family2026.01 | 43.2 | |
| Molmo2-8BCategory=Fully open2026.03 | 43.2 | |
| Gemini 2.5 ProAccess Type=API call only2026.01 | 42.1 | |
| Gemini 2.5 ProCategory=API call only2026.03 | 42.1 | |
| MolmoPoint-8BCategory=MolmoPoint2026.03 | 40.3 | |
| Molmo2-O-7BAccess Type=Molmo2 family2026.01 | 40.1 | |
| Molmo2-O-7BCategory=Fully open2026.03 | 40.1 | |
| Molmo2-4BAccess Type=Molmo2 family2026.01 | 39.9 | |
| Molmo2-4BCategory=Fully open2026.03 | 39.9 | |
| Gemini 3 ProAccess Type=API call only2026.01 | 36 | |
| Gemini 3 ProCategory=API call only2026.03 | 36 | |
| MiniCPM-V-4.5-8BAccess Type=Open weights only2026.01 | 29.3 | |
| MiniCPM-V-4.5-8BCategory=Open weights2026.03 | 29.3 | |
| Qwen3-VL-8BAccess Type=Open weights only2026.01 | 26.7 | |
| Qwen3-VL-8BCategory=Open weights2026.03 | 26.7 | |
| Claude Sonnet 4.5Access Type=API call only2026.01 | 26 | |
| Claude Sonnet 4.5Category=API call only2026.03 | 26 | |
| Keye-VL-1.5-8BAccess Type=Open weights only2026.01 | 25.4 | |
| Keye-VL-1.5-8BCategory=Open weights2026.03 | 25.4 | |
| Qwen3-VL-4BAccess Type=Open weights only2026.01 | 25.2 | |
| Qwen3-VL-4BCategory=Open weights2026.03 | 25.2 | |
| Eagle2.5-8BAccess Type=Open weights only2026.01 | 22.8 | |
| Eagle2.5-8BCategory=Open weights2026.03 | 22.8 | |
| LLaVA-Video-7BAccess Type=Open models2026.01 | 19.9 | |
| LLaVA-Video-7BCategory=Fully open2026.03 | 19.9 | |
| GLM-4.1V-9BAccess Type=Open weights only2026.01 | 18.4 | |
| GLM-4.1V-9BCategory=Open weights2026.03 | 18.4 | |
| VideoChat-Flash-7BAccess Type=Open models2026.01 | 14.8 | |
| VideoChat-Flash-7BCategory=Fully open2026.03 | 14.8 | |
| PLM-3BAccess Type=Open models2026.01 | 12.3 | |
| PLM-3BCategory=Fully open2026.03 | 12.3 | |
| PLM-8BAccess Type=Open models2026.01 | 10.9 | |
| PLM-8BCategory=Fully open2026.03 | 10.9 | |
| InternVL3.5-8BAccess Type=Open weights only2026.01 | 7.8 | |
| InternVL3.5-8BCategory=Open weights2026.03 | 7.8 | |
| InternVL3.5-4BAccess Type=Open weights only2026.01 | 7.7 | |
| InternVL3.5-4BCategory=Open weights2026.03 | 7.7 |