Skip to content

Dataset · lmms-lab-encoder

Evaluation dataset for MME benchmark

lmms-lab-encoder/MME

Multimodal image-text dataset for testing vision-language models (VLMs).

What's inside

This dataset serves as an evaluation split for the MME benchmark and includes between 1 and 10 thousand text and image examples. The language and detailed column structure are not specified in the card.

What you can do with it

  • Evaluate basic visual recognition and analysis capabilities of vision-language models (VLMs)
  • Perform regression testing on multimodal models after fine-tuning
Size
1K–10K rows
Downloads this month
25.6K
Likes
37

How to load it

from datasets import load_dataset

ds = load_dataset("lmms-lab-encoder/MME")