OpenCodePapers

visual-document-retrieval-on-vidore-v3

Document UnderstandingVisual Document Retrieval
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeNDCG@10NDCG@5ModelNameReleaseDate
EVIE: High-Performance Multilingual Visual Document Retrieval with Matryoshka Embeddings and Token Compression✓ Link66.7564.16EVIE-8B2026-09-07
Nemotron ColEmbed V2: Top-Performing Late Interaction embedding models for Visual Document Retrieval63.53Nemotron ColEmbed VL 8B V22026-02-03
MTEB Leaderboard57.64ColNomic 7B2025-04-02
MTEB Leaderboard✓ Link57.54jina-embeddings-v42025-06-23
Nemotron ColEmbed V2: Top-Performing Late Interaction embedding models for Visual Document Retrieval57.07Llama NemoRetriever ColEmbed 3B v12026-02-03
MTEB Leaderboard56.4ColNomic 3B2025-04-02
NeoMME: A Single-Tower Multimodal-Native Multilingual Foundation Encoder for Efficient Fine-Tuning and Inference55.653.29NeoMME-800M-Retriever (late interaction)2026-08-31
Nemotron ColEmbed V2: Top-Performing Late Interaction embedding models for Visual Document Retrieval55.48Llama NemoRetriever ColEmbed 1B v12026-02-03
MTEB Leaderboard✓ Link52.44ColQwen2.5 v0.22024-06-27
MINER: Mining Multimodal Internal Representation for Efficient Retrieval49.7MINER-Jina2026-05-07
jina-embeddings-v5-omni: Text-Geometry-Preserving Multimodal Embeddings via Frozen-Tower Composition✓ Link45.71jina-embeddings-v5-omni-small2026-05-08
CausalEmbed: Auto-Regressive Multi-Vector Generation in Latent Space for Visual Document Embedding✓ Link45.5542.6CausalQwen2026-01-29
NanoVDR: Distilling a 2B Vision-Language Retriever into a 70M Text-Only Encoder for Visual Document Retrieval46.5NanoVDR-S-Multi2026-03-13
Hydra: Unifying Document Retrieval and Generation in a Single Vision-Language Model56.39Hydra-4B2026-03-30
DistilVDR: A Compact End-to-End Visual Document Retriever via Dual-Student Distillation✓ Link47.07DistilVDR-HiRes2026-08-11
ColNanoVDR: Document-Free Query Distillation for Multi-Vector Visual Document Retrieval via Optimal Transport✓ Link60.1ColNanoVDR (from ColVec1.1-8b)2026-09-28
MTEB Leaderboard✓ Link44.18ColQwen2 v1.02024-06-27
MTEB Leaderboard✓ Link42.95ColPali v1.32024-06-27
MTEB Leaderboard✓ Link26.12ColModernVBERT2025-10-01
MTEB Leaderboard✓ Link17.68135417SigLIP Large Patch16 3842023-03-27
MTEB Leaderboard✓ Link17.2255SigLIP SO400M Patch14 3842023-03-27
MTEB Leaderboard✓ Link15.41BiModernVBERT2025-10-01
MTEB Leaderboard✓ Link15.17822917SigLIP Base Patch16 5122023-03-27
MTEB Leaderboard✓ Link14.96ModernVBERT-embed2025-10-01
MTEB Leaderboard✓ Link12.3615625SigLIP Base Patch16 3842023-03-27
MTEB Leaderboard✓ Link9.02CLIP ViT-L/142021-02-26
MTEB Leaderboard✓ Link7.7263125SigLIP Base Patch16 2562023-03-27
MTEB Leaderboard✓ Link7.236SigLIP Base Patch16 256 Multilingual2023-03-27
MTEB Leaderboard✓ Link6.31527083SigLIP Base Patch16 2242023-03-27
MTEB Leaderboard✓ Link5.25CLIP ViT-B/162021-02-26
MTEB Leaderboard✓ Link2.68CLIP ViT-B/322021-02-26
Check out how to contribute new results. Then edit this file on GitLab or GitHub.