Skip to content

ViDoRe v2

Visual Document Retrieval benchmark version 2.

Overview

Field Value
Modality Multimodal (Images)
Generation GT No
HF Repository vidorev2-dumps

Description

ViDoRe v2 is an updated version of the Visual Document Retrieval benchmark with improved annotations and additional document types.

Download

autorag-research data restore vidorev2 <dataset_name>_<embedding_model>

Ingest from Source

autorag-research ingest --name=vidorev2 --extra dataset-name=<name> --embedding-model=colpali

Best For

  • Visual document understanding
  • PDF page retrieval
  • Multimodal embedding evaluation