List your company
ResearchFormerly WizWand

Video Reasoning on MMMU Video

84.6AccuracyGPT-5-thinking
21.57637.93854.370.662Nov 24, 2025
Updated 8mo ago

Evaluation results

MethodLinks
GPT-5-thinkingModel Category=Closed-source Large-sized MLLM2025.1184.6
Gemini 2.5 ProModel Category=Closed-source Large-sized MLLM2025.1183.6
OpenAI O3Model Category=Closed-source Large-sized MLLM2025.1183.3
Seed 1.5VLModel Category=Closed-source Large-sized MLLM2025.1181.4
Qwen3-VL-235B-ThinkingModel Category=Open-source Large-sized MLLM2025.1180
VideoChat-M1Model Category=Ours, Number of Parameters=37B2025.1180
Qwen3-VL-235B-InstructModel Category=Open-source Large-sized MLLM2025.1174.7
Qwen3-VL-8B-InstructModel Category=Open-source Medium-sized MLLM2025.1165.3
GPT-4oModel Category=Closed-source Large-sized MLLM2025.1161.2
Gemini 1.5 ProModel Category=Closed-source Large-sized MLLM2025.1153.9
VideoChat-R1.5-7BModel Category=Open-source Medium-sized MLLM2025.1151.4
Aria-28BModel Category=Open-source Large-sized MLLM2025.1150.8
LLAVA-Video-72BModel Category=Open-source Large-sized MLLM2025.1149.7
LLaVA-ov-72BModel Category=Open-source Large-sized MLLM2025.1148.3
LLaVA-Video-7BModel Category=Open-source Medium-sized MLLM2025.1136.1
LongVA-7BModel Category=Open-source Medium-sized MLLM2025.1124