Share your thoughts, 1 month free Claude Pro on usSee more

Social Video Reasoning on VAGUE (test)

77.4Accuracy

Gemini-3.0-Flash

Updated 1mo ago

Evaluation Results

Method	Links
Gemini-3.0-Flash 2025.07		77.4
Gemini-3.0-Flash 2025.07		73.05
Gemini-3.0-Flash 2025.07		72.27
Gemini-2.5-Pro 2025.07		67.62
GPT-4o 2025.07		67.43
Claude-3.5-Sonnet 2025.07		67.3
Claude-3.5-Sonnet 2025.07		62.92
OpenAI o1-full 2025.07		62.57
GPT-5.2-High 2025.07		62.02
GPT-4o 2025.07		61.6
OpenAI o1-full 2025.07		61.59
GPT-4o 2025.07		60
Gemini-3.0-Flash 2025.07		60
GPT-5.2-High 2025.07		59.99
OpenAI o1-full 2025.07		59.82
Claude-3.5-Sonnet 2025.07		59.52
GPT-5.2-High 2025.07		59.21
Gemini-2.5-Pro 2025.07		58.32
LLaVA-OneVision-7B 2025.07		54.65
LLaVA-OneVision-7B 2025.07		53.31
Gemini-2.5-Pro 2025.07		53.25
Claude-3.5-Sonnet 2025.07		52.02
Idefics3-8B 2025.07		50.98
LLaVA-v1.6-7B 2025.07		50.21
GPT-4o 2025.07		50.12
Molmo-7B 2025.07		50
LLaVA-OneVision-7B 2025.07		49.73
OpenAI o1-full 2025.07		48.84
Idefics3-8B 2025.07		48.54
Idefics3-8B 2025.07		48.24
Molmo-7B 2025.07		48
LLaVA-v1.6-7B 2025.07		47.44
GPT-5.2-High 2025.07		47.31
Gemini-2.5-Pro 2025.07		46.7
Molmo-7B 2025.07		46
LLaVA-v1.6-7B 2025.07		46
Qwen2-VL-7B 2025.07		44.42
LLaVA-OneVision-7B 2025.07		44.3
OpenAI o3-mini 2025.07		43.64
Idefics3-8B 2025.07		42.43
OpenAI o3-mini 2025.07		40.21
Molmo-7B 2025.07		40.19
LLaVA-v1.6-7B 2025.07		40.1
Qwen2-VL-7B 2025.07		38.52
Qwen2-VL-7B 2025.07		38.49
OpenAI o3-mini 2025.07		38.15
OpenAI o3-mini 2025.07		34.43
Qwen2-VL-7B 2025.07		34.02