Skip to content

Dataset · mjuicem

StreamingBench: video streaming analysis benchmark

mjuicem/StreamingBench

English video QA dataset for evaluating multimodal neural network video understanding in real time.

What's inside

A benchmark of 1k to 10k instances (videos, images, and English text questions) created to test Multimodal Large Language Models (MLLMs) in streaming video mode. It evaluates how well neural networks handle real-time event analysis, including proactive responses without direct prompting.

What you can do with it

  • Evaluating and comparing MLLMs in streaming Video-QA tasks
  • Testing video analytics systems requiring real-time response
Tasks
Question answering
Languages
English
Size
1K–10K rows
Downloads this month
38K
Likes
14

How to load it

from datasets import load_dataset

ds = load_dataset("mjuicem/StreamingBench")