List your company

Video Understanding on TempCompass MCQ (test)

82.8AccuracyGemini 3 Pro
44.00854.07964.1574.221Jan 15, 2026Jan 27, 2026Feb 8, 2026Feb 21, 2026Mar 5, 2026Mar 17, 2026Mar 30, 2026

Evaluation results

MethodLinks
Gemini 3 ProAccess Type=API call only2026.0182.8
Gemini 3 Protype=Proprietary2026.0282.8
Gemini 3 ProCategory=API call only2026.0382.8
Gemini 2.5 ProAccess Type=API call only2026.0181.9
Gemini 2.5 Protype=Proprietary2026.0281.9
Gemini 2.5 ProCategory=API call only2026.0381.9
GPT-5Access Type=API call only2026.0180.4
GPT-5type=Proprietary2026.0280.4
GPT-5Category=API call only2026.0380.4
Gemini 2.5 FlashAccess Type=API call only2026.0180.2
Gemini 2.5 FlashCategory=API call only2026.0380.2
Keye-VL-1.5-8BAccess Type=Open weights only2026.0175.5
Keye-VL-1.5-8BCategory=Open weights2026.0375.5
GPT-5 miniAccess Type=API call only2026.0174.9
GPT-5 miniCategory=API call only2026.0374.9
Eagle2.5-8BAccess Type=Open weights only2026.0174.4
Eagle2.5-8BCategory=Open weights2026.0374.4
MolmoPoint-8BCategory=MolmoPoint2026.0374.4
Qwen3-VL-8BAccess Type=Open weights only2026.0174.3
Qwen3-VL-8BCategory=Open weights2026.0374.3
Molmo2-8BAccess Type=Molmo2 family2026.0173.4
Molmo2-8BCategory=Fully open2026.0373.4
Molmo2-O-7BAccess Type=Molmo2 family2026.0173
Molmo2-O-7BCategory=Fully open2026.0373
Claude Sonnet 4.5Access Type=API call only2026.0172.8
Molmo2-4BAccess Type=Molmo2 family2026.0172.8
Claude Sonnet 4.5type=Proprietary2026.0272.8
Claude Sonnet 4.5Category=API call only2026.0372.8
Molmo2-4BCategory=Fully open2026.0372.8
MiniCPM-V-4.5-8BAccess Type=Open weights only2026.0172.7
PLM-8BAccess Type=Open models2026.0172.7
MiniCPM-V-4.5-8BCategory=Open weights2026.0372.7
PLM-8BCategory=Fully open2026.0372.7
GLM-4.1V-9BAccess Type=Open weights only2026.0172.3
GLM-4.1V-9BCategory=Open weights2026.0372.3
Qwen3-VL-4BAccess Type=Open weights only2026.0170.8
Qwen3-VL-4BCategory=Open weights2026.0370.8
InternVL3.5-8BAccess Type=Open weights only2026.0170.3
InternVL3.5-8BCategory=Open weights2026.0370.3
VideoChat-Flash-7BAccess Type=Open models2026.0169.4
VideoChat-Flash-7BCategory=Fully open2026.0369.4
PLM-3BAccess Type=Open models2026.0169.3
PLM-3BCategory=Fully open2026.0369.3
InternVL3.5-4BAccess Type=Open weights only2026.0168.8
InternVL3.5-4BCategory=Open weights2026.0368.8
CoPE-VideoLMnumber of parameters=7B2026.0268.4
IXC-2.5-7Bnumber of parameters=7B2026.0267.1
LLaVA-Video-7BAccess Type=Open models2026.0166.6
LLaVA-Video-7Bnumber of parameters=7B2026.0266.6
LLaVA-Video-7BCategory=Fully open2026.0366.6
InternVL2-8Bnumber of parameters=8B2026.0265.3
Apollo-7Bnumber of parameters=7B2026.0264.9
LLaVA-OV-7Bnumber of parameters=7B2026.0264.8
LongVA-7Bnumber of parameters=7B2026.0256.9
VideoChat2-7Bnumber of parameters=7B2026.0245.5